Contents lists available at ScienceDirect

Industrial Marketing Management

journal homepage: www.elsevier.com/locate/indmarman

Research paper

Boosting sales force morale in highly dynamic, complex markets: The role ofjob resources

Nikolaos G. Panagopoulosa,⁎, Bryan Hochsteinb, Thomas L. Bakerb, Michael A. Pimentelb

aOhio University, Department of Marketing, 238 Copeland Hall, Athens, OH 45701, United Statesb The University of Alabama, Department of Marketing, Box 870225, Tuscaloosa, AL 35487, United States

A R T I C L E I N F O

Keywords:MoraleSales forceProductivityTurnoverJD-R theory

A B S T R A C T

Sales force morale constitutes an important managerial topic that is often linked to key outcomes such as salesforce turnover and productivity. Unfortunately, however, scholarly work in this area is strikingly limited.Accordingly, the goal of this study is to provide a first rigorous assessment of the role of morale in a sales context.Drawing on Job Demands-Resource (JD-R) as our theoretical lens and using a unique dataset that includesresponses from three sources (i.e., sales managers, salespeople, and secondary objective data) from 81 companiesover two time periods, our study makes several contributions. First, we offer a conceptualization of sales forcemorale and thus advance this timely and managerially relevant topic in a JD-R setting. Second, we show thenegative impact of market demands (i.e., customer purchase complexity and market dynamism) on sales forcemorale. Third, the findings highlight the positive impact of morale on sales force turnover and productivity.Fourth, results show that two job resources attenuate the negative impact of market demands on sales forcemorale (i.e., sales capabilities training sales unit's cross-functional cooperation). Surprisingly, however, we findthat a third job resource – that is, a firm's product portfolio depth – actually accentuates, rather than attenuates,the negative effects of market demands on sales force morale. We conclude by discussing the theoretical andmanagerial implications of our work and by elaborating on exciting avenues for future research in the area.

1. Introduction

Improving sales force morale is widely regarded among practi-tioners as a valuable strategy that can significantly enhance key out-comes such as job performance and turnover (Martin, 2015). Super-normal turnover at companies such as Groupon, for example, has beenattributed to low levels of salespeople's morale (Ovide, 2012). Notsurprisingly, therefore, proactive companies, such as John Deere, areinvesting substantial resources to systematically learn about, measure,and manage employee morale (Power, 2016). Despite this level ofpractitioners' interest, scholars have not demonstrated an equal amountof attention in the notion of sales force morale. As shown in Table 1, thetopic of sales force morale has been the subject of very limited researchin the extant marketing literature. As such, two important research gapsremain.

First, the concept of sales force morale has not been the direct focusof much research in the extant marketing literature, with only tan-gential reference to the concept and without providing a clear defini-tion or how it differs from other constructs, which seem to be over-shadowing the concept of sales force morale (see Table 1). For instance,

morale has been equated to and used interchangeably with the con-structs of motivation (Cotham III, 1968), general feeling states and at-titudes (Mantel, 2005), or satisfaction (Churchill, Ford, & Walker,1976). Given this lack of attention, prior research offers little specificguidance concerning ways managers can employ to manage sales forcemorale.

Second, although prior research on morale outside marketing hasprovided useful insights on what morale is in a general employee set-ting (e.g., Chang, Rosen, & Levy, 2009; Subramony, Krause, Norton, &Burns, 2008), at least two realities of the modern sales position requireattention to how morale manifests and functions as well as to what areits antecedents, boundary conditions, and consequences in a sales forcecontext. First, many salespeople work in physical, social, and psycho-logical isolation from the firm for which they work (Dubinsky, Howell,Ingram, & Bellenger, 1986; Ingram, LaForge, Locander, MacKenzie, &Podsakoff, 2005) thereby making activity less visible to management.Reduced proximity to management is further amplified today due tomany firms' initiatives for salespeople to work in “virtual offices” wheresalespeople work from remote locations with fewer chances for inter-action with their supervisors (Mulki & Jaramillo, 2011). Isolation from

https://doi.org/10.1016/j.indmarman.2018.06.001Received 24 November 2017; Received in revised form 1 April 2018; Accepted 1 June 2018

⁎ Corresponding author.E-mail addresses: npanag@ohio.edu (N.G. Panagopoulos), bwhochstein@cba.ua.edu (B. Hochstein), tbaker@cba.ua.edu (T.L. Baker), mapimentel@cba.ua.edu (M.A. Pimentel).

Industrial Marketing Management xxx (xxxx) xxx–xxx

0019-8501/ © 2018 Elsevier Inc. All rights reserved.

Please cite this article as: Panagopoulos, N.G., Industrial Marketing Management (2018), https://doi.org/10.1016/j.indmarman.2018.06.001

Table1

Rep

resentativeresearch

onmorale.

Stud

yFo

cusof

stud

yDefi

nition

Ope

ration

alde

finition

Datasource(s)

Type

Other

commen

ts

Baeh

ran

dRen

ck(195

8)Aim

ofinve

stigationwas

tode

fine

employ

eemoralean

ditsdimen

sion

s.Doe

sno

texplicitly

define

employ

eemorale

Theau

thorsde

scribe

employ

eemoraleas

afunc

tion

ofintegrationwiththe

orga

nization

,jobsatisfaction

,immed

iate

supe

rvision,

andfriend

linessan

dco

operationof

co-w

orke

rs.

Indu

strial

employ

ees

Empirical

Theau

thorsdraw

on‘needs

theo

ries’to

deve

lopamulti-dim

ension

alscaleof

employ

eemorale.

Brittan

dDickinson

(200

6)Ev

alua

testherole

ofmorale,

compa

redto

depression

,asapo

sitive

psycho

logy

construc

t.

“Aservicemem

ber's

leve

lof

motivationan

den

thusiasm

forachiev

ingmission

success”

(p.1

62)

Four-item

scalewhe

resoldiers

ratedtheir

leve

lof

person

almorale,

motivation,

energy

,and

drive.

U.S.m

ilitary

that

were

deploy

edin

ape

acek

eeping

mission

Empirical

Moralewas

foun

dto

bepo

sitive

lyrelatedto

enga

gemen

tin

meaning

fulworkan

dco

nfide

ncein

unitfunc

tion

ingan

dlead

ership.

Cam

pbellan

dTy

ler

(195

7)Th

isstud

yattemptsto

prov

ideva

lidityfor

moralesurvey

s.A

grou

p'sav

erag

efeelingof

conten

tmen

tor

satisfaction

abou

tthemajor

aspe

ctsof

theworksituation.

Thestud

yutilizesa30

-item

moralesurvey

.Sa

mple1)

Employ

eesof

insuranc

eco

mpa

nies

and

sample2)

men

inana

vel

squa

dron

of10

subm

arines.

Empirical

Theau

thorsfind

that

secu

ring

work-grou

prepu

tation

salon

gwithmoralesurvey

sprov

ides

ach

eckforco

nstruc

tva

lidity.

Cha

ng,R

osen

,and

Levy

(200

9)Und

erstan

ding

how

employ

ees'

percep

tion

sof

orga

nization

alpo

litics

influe

nceindividu

al-le

velworkattitude

san

dbe

havior.

Agg

rega

tedge

neral

employ

eeattitude

s.Agg

rega

tedjobsatisfaction

andaff

ective

commitmen

t.Meta-an

alysis

Empirical

Employ

eemoralepe

rcep

tion

sof

orga

nization

alpo

liticsto

performan

ce.

Chu

rchill,

Ford,a

ndWalke

r(197

6)Th

eeff

ectof

orga

nization

alclim

ate

variab

leson

jobsatisfaction

.NA

NA

Salesperson

Empirical

Moraleis

equa

tedto

andused

intercha

ngeablywithjobsatisfaction

Cotha

mIII(196

8)Th

erelation

ship

betw

eensalesjob

satisfaction

andpe

rforman

ce.

NA

NA

Salesperson

Empirical

Moraleis

equa

tedto

andused

intercha

ngeablywithmotivationan

djob

satisfaction

Deeter-Schm

elz,

Goe

bel,an

dKen

nedy

(200

8)

Thech

aracteristicsan

dco

nseq

uenc

esof

effective

salesman

agers.

NA

NA

Salesperson,

Salesman

ager

Empirical

Positive

moraleis

describe

dby

man

agersas

anou

tcom

eof

effective

man

agem

ent

Dickson

,Fa

rris,a

ndVerbe

ke(200

1)A

taxo

nomyof

feed

back

regu

larities

toim

prov

eman

agers'strategicthinking

.NA

NA

NA

Con

ceptua

lLo

wem

ploy

eemorale(not

define

d)is

describe

das

having

aco

ntag

ioneff

ectthat

nega

tive

lyeff

ects

performan

ce.

Man

tel(200

5)Con

textua

lan

dpe

rson

alinflue

nces

onsalespersonethicalde

cision

-mak

ing.

Moraleis

describe

dge

nerally

toinvo

lvefeeling

states

andattitude

s.

NA

Salespeo

ple

Empirical

Briefdiscussion

onho

wmoraleis

interpreted

inco

nsistently

withinan

dacross

literature

domains,b

utno

clearde

finition

.Rosen

,Lev

y,an

dHall(200

6)Su

ggests

that

orga

nization

alen

vironm

ents

characterizedwithhigh

leve

lsof

inform

alsupe

rvisor

supp

ortan

dco

worke

rfeed

back

areassociated

with

lower

leve

lsof

perceive

dorga

nization

alpo

litics.

Moralestem

sfrom

individu

al,bu

tman

ifests

atgrou

pleve

l

Theau

thorsop

erationa

lized

moraleas

ahigh

erorde

rreflective

construc

twith

dimen

sion

sof

jobsatisfaction

andaff

ective

commitmen

t

College

stud

ents

and

supe

rvisors

Empirical

Moralewas

foun

dto

med

iate

therelation

ship

betw

eenpo

liticsan

dmultidimen

sion

alpe

rforman

ce.

Subram

ony,

Krause,

Norton,

and

Burns(200

8)

Und

erstan

ding

how

inve

stmen

tsin

human

resourcesin

theform

ofco

mpe

titive

paycanincrease

orga

nization

alpe

rforman

ce.

Employ

ees'co

llective

attitude

stowardthe

orga

nization

.

Theau

thorsop

erationa

lizemoraleas

athree-item

scalerepresen

ting

employ

ees'

satisfaction

withtheirem

ploy

eran

dloya

lty

towardtheorga

nization

.

Rep

resentativesampleof

U.S.

workforce

Empirical

Moraleis

foun

dto

med

iate

therelation

ship

betw

eenpe

rcep

tion

sof

compe

titive

payan

dfuture

custom

ersatisfaction

.

Tice

(199

7)Fram

eworkforclassifyingan

dim

plem

enting

caps

onsalesforce

compe

nsationplan

s

NA

NA

N/N

ACon

ceptua

lSa

lesforcemoraleistheo

rizedto

beda

mag

edto

varyingde

greesof

seve

rity

basedon

the

metho

dan

dtimingof

chan

gesto

salesforce

compe

nsationplan

s.Current

Stud

yHow

salesman

agerscanutilize

firm

resourcesto

attenu

atethene

gative

deman

dsplaced

onsalesforcemorale

Thesalesforce'sco

llective

attitude

stowardmajor

aspe

ctsof

thejoban

dorga

nization

Measuredas

asalesforce-leve

l,seco

nd-

orde

rreflective

construc

twiththree

dimen

sion

s:aff

ective

orga

nization

alco

mmitmen

t,satisfaction

withthejob,

and

satisfaction

withco

mpa

nypo

licy/

proc

edures.

Salesperson,

Salesman

ager

Empirical

Using

JD-R

theo

ry,the

curren

tstud

yis

the

firstin

marke

ting

toco

ncep

tualizean

dem

pirically

exam

inesalesforcemorale,

includ

ingitsan

tecede

nts,

outcom

es,a

ndman

agerially

controlla

blemod

erators.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

2

management renders salespeople less readily susceptible to leadership,motivation, and coaching interventions that may result from manager-salesperson interactions. Fewer outlets of visibility and managementintervention provide greater occasion for salespeople to feel isolated,which can lead to decreased levels of job satisfaction and organizationalcommitment (Mulki, Locander, Marshall, Harris, & Hensel, 2008).Second, there is universal consensus among sales academics and prac-titioners that the sales job is becoming increasingly complex and dy-namic (Jones, Brown, Zoltners, & Weitz, 2005; Plouffe, Bolander, Cote,& Hochstein, 2016; Schmitz & Ganesan, 2014). Powerful forces stem-ming from evolutions in technology, customer demands, and new formsof competition all create a new “culture” that requires salespeople toadapt quickly and effectively to the velocity with which companiesimplement new customer strategies, launch new products, and redefinetheir selling models. This context, however, imposes “escalating de-mands and expectations on salespeople in virtually all industries”(Ingram, LaForge, & Schwepker, 2011, p. 253), thereby creating con-ditions conducive to lower sales force morale.

Against this background, our study makes three novel contributionsto marketing research. First, we offer a rigorous investigation of the roleof morale in a sales context. Specifically, we draw from research inmanagement (e.g., Chang, Rosen, & Levy, 2009; Subramony, Krause,Norton, & Burns, 2008) to define sales force morale as the sales force'scollective attitudes toward major aspects of the job and organization(see Table 1). We believe that a better understanding of sales forcemorale provides unique insights to both managers and academicians.

Second, we shed more light on the antecedents, moderating condi-tions, and consequences of sales force morale by employing job-demandresource theory (JD-R) as our theoretical lens. In essence, JD-R theorysuggests that employee job strain, and thus reduced levels of perfor-mance, occurs when there is an imbalance between demands on em-ployees and the resources available to respond to those demands.Accordingly, we investigate the negative impact of two salient demandsinherent in the external market environment – that is, customer pur-chase complexity and market dynamism on sales force morale (seeFig. 1). Both customer complexity and market dynamism have receivedrecent research attention because they are key attributes of the chan-ging sales environment (e.g., Plouffe, Bolander, Cote, & Hochstein,2016; Schmitz & Ganesan, 2014). Furthermore, JD-R theory suggeststhat increases in job resources – that is, “tools” provided by the firm tohelp the sales force successfully manage increased demands – can“buffer” the negative impacts of job demands (Bakker & Demerouti,2014). Consistent with this notion, we also explore three job resourcesthat may attenuate the negative impact of market demands on morale(i.e., sales capabilities training, firm's product portfolio depth, and salesunit's cross-functional cooperation). Finally, we empirically test asser-tions of the beneficial effects of sales force morale on key outcomes. Inparticular, we investigate the extent to which sales force morale reducessales force turnover and increases sales force productivity.

Finally, the third contribution of our work comes from the fact thatwhile JD-R theory has to date been used at the individual-level (Bakker,Demerouti, & Verbeke, 2004), here we extend the theory to the salesforce level, thus answering recent calls for research at the firm-level(Bakker & Demerouti, 2014, 2018). This extension has been advocatedbecause it is currently unknown how the elements of JD-R will operateat different levels. In fact, recent work suggests that resources thatbuffer the effects of demands at the individual-level may actually ex-acerbate relationships at the firm-level (Bakker & Demerouti, 2018). Tothis end, we empirically test the notion that firms can “proactively re-design” jobs at an organizational (e.g., sales force) level in an effort tooffset the negative effects of environmental demands. Fig. 2 illustratesthe structure and levels of the study variables. Specifically, all con-structs in our conceptual model (Fig. 1) are at the sales force-level. Inaddition, consistent with the prescriptions of compositional models inmultilevel theory (Chan, 1998; Kozlowski & Klein, 2000), which wediscuss subsequently, sales force morale originates at the individual

salesperson level but manifests as a collective phenomenon at the salesforce-level.

Our contributions derive in large part from our ability to employ aunique data set that combines data collected from three sources (i.e.,sales managers, salespeople, and secondary objective data) across twotime periods for 81 firms. Accordingly, our results provide sales ex-ecutives with actionable implications for managing sales force morale.Specifically, the results suggest that market demands hurt sales forcemorale, whereas offering training that improves sales force capabilitiesor creating a cooperative relationship between the sales unit andmarketing or R&D functions buffer the negative influence of marketdemands. Interestingly, we also find that a deeper product portfoliomagnifies the negative effects of a demanding selling context on salesforce morale. Finally, our results reveal that boosting sales force moralematters significantly. Specifically, we find that an increase of morale byone point on a 5-point scale improves sales force productivity by€226,834 of operating revenues per salesperson, while lowering turn-over rate by 5%.

The remainder of the paper is organized as follows. First, we ela-borate on the construct of sales force morale as well as on the JD-Rtheory upon which we build on. Next, we delve into the conceptuallogic of our model and hypotheses, which is followed by details con-cerning our dataset, methods employed, and results. Finally, we discussour findings along with limitations and future research directions.

2. Theoretical background

2.1. Sales force morale

Despite the lack of attention on morale in the marketing literature,the concept of morale is well established in the organizational domain(e.g., Peterson, Park, & Sweeney, 2008). Specifically, there is agreementin the extant literature that morale reflects an aggregation of attitudinalvariables that stem from the individual employee but manifest at thegroup or organization level (Rosen, Levy, & Hall, 2006; Subramony,Krause, Norton, & Burns, 2008). Accordingly, morale refers to “em-ployees' collective attitudes toward the organization” (Subramony,Krause, Norton, & Burns, 2008, p. 780). Furthermore, given that jobsatisfaction and affective organizational commitment constitute the twomost important attitudes in organizational research (see Harrison,Newman, & Roth, 2006), prior work has employed these two attitudinalconstructs to define morale at the group/organizational level (e.g.,Chang, Rosen, & Levy, 2009; Harrison, Newman, & Roth, 2006; Rosen,Levy, & Hall, 2006; Subramony, Krause, Norton, & Burns, 2008).

As mentioned previously, morale can be thought of as the collectiveattitudes of individuals, representing how positive the group feels as awhole about their shared situation (e.g., Subramony, Krause, Norton, &Burns, 2008). Stated differently, a key aspect of morale is that, incontrast to individual level measures of satisfaction or commitment, itrefers to shared attitudes of employees within a group (Britt &Dickinson, 2006). This view on morale is consistent with the prescrip-tions of compositional models in multilevel theory (Chan, 1998) as wellas social information processing theory (Salancik & Pfeffer, 1978).Specifically, shared attitudes emerge and manifest at the organizationallevel through a bottom-up process involving social interaction, devel-opment of common experiences, information exchange, and inter-dependencies in work processes among employees (Kozlowski & Klein,2000; Schulte, Ostroff, Shmulyian, & Kinicki, 2009). Specifically, due tocommon influences, such as sharing the same leader and practices, aswell as social interaction and communication, employees in the sameunit focus on similar aspects of the organization, thus cognitively andaffectively evaluating shared job experiences until they gradually reachshared level of feelings toward the organization (Whitman, Van Rooy, &Viswesvaran, 2010). This process of formation of shared attitudes is akey characteristic of sales force key because it explicitly acknowledgesthat salespeople may have norms of cooperation and collaboration,

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

3

which help enhance the sales unit's productivity above and beyond theinfluence of individual employee attitudes (Koys, 2001). Our con-ceptualization of sales force morale draws from this line of research andthus it allows us to approach morale at the organizational (i.e., salesforce) rather than the individual-level. This focus on the collectivenature of sales force morale is desirable and in line with recent salesliterature which has urged researchers to examine the drivers of per-formance at the sales force level (Cron, 2017; Verbeke, Dietz, &Verwaal, 2011).

While morale reflects employees' attitudes toward major aspects ofthe job and organization, it is important to distinguish morale fromother related constructs, such as psychological climate as well as or-ganizational culture. First, psychological climate refers to meaningfulperceptions an individual employee develops of his/her work environ-ment with regards to its structure, processes, and events (Schneider,1975). These perceptions help an employee determine how beneficialor detrimental the work environment is to his/her well-being (James,James, & Ashe, 1990) and have been found to have a significant re-lationship with employee work attitudes, motivation, and performance(e.g., Hartmann & Rutherford, 2015). In contrast, sales force moralerepresents attitudes rather than perceptions, which can lead to theformation of attitudes. Indeed, climate has been found to be an ante-cedent of sales force morale (Churchill, Ford, & Walker, 1976). In ad-dition, while psychological climate refers to individual perceptions ofand the meanings they assign to their environment (Dickson, Resick, &Hanges, 2006), morale refers to organizational-level attitudes(Subramony, Krause, Norton, & Burns, 2008). Second, organizationalculture refers to the set of norms, values, and beliefs that define the howbusiness should be conducted within the organization (Barney, 1986).The culture of an organization dictates appropriate employee behavior

and prescribes the ways problems are to be addressed (Schein, 1992). Inother words, culture represents why things happen the way they do inany organization and is often difficult to change (Deshpande & WebsterJr, 1989; Schneider & Rentsch, 1987). In contrast, sales force moralereflects salespeople's collective attitudes that can change more easilyand frequently. For example, managers can implement new incentiveprograms that benefit and increase sales force morale.

Against this background, we define sales force morale as the salesforce's collective attitudes toward major aspects of the job and orga-nization (Table 1). Accordingly, we operationalize sales force morale asa multidimensional concept comprising collective salespeople's atti-tudes of (a) affective organizational commitment, (b) satisfaction withthe job, and (c) satisfaction with company policies and procedures (seeMethods section for measurement details).

Job Demands-Resource Theory.Our conceptual model is directly informed by JD-R theory. First,

employee behavior and performance are a function of job character-istics, which can be modeled using two different categories: job de-mands and job resources. Job demands are “physical, psychological,social, or organizational aspects of the job that require sustained phy-sical or psychological (cognitive or emotional) effort or skills and aretherefore associated with certain physical and/or psychological costs”(Bakker & Demerouti, 2007, p. 312). The JD-R model has typically beenused to describe the costs of job demands in terms of negative phy-siological and/or psychological outcomes (i.e., increased job demandsincrease negative outcomes). For example, work pressure and/oremotional demands are expected to result in costs, such as burnoutderived from the consistent effort required to meet the demands(Bakker, Demerouti, & Verbeke, 2004). As mentioned previously, jobdemands are represented by two market-related factors in our study:

Fig. 1. Conceptual model.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

4

customer purchase complexity and market dynamism.Second, JD-R theory also suggests that when facing job demands,

employees will draw on the personal and organizational resourcesavailable in an effort to buffer the negative effects job demands exert onemployee outcomes such as engagement or performance (Bakker &Demerouti, 2007). Job resources are physical, psychological, social, ororganizational components of a job that function to help salespeopleachieve work goals, reduce job demands (and their costs), and/or sti-mulate learning (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001).When job resources are properly implemented, they can be motivatorsthat improve work engagement, reduce cynicism, and improve perfor-mance (Bakker & Demerouti, 2007). Job resources typically reduce thenegative effects of a job demand by giving a worker the right tools to dothe job. For example, in a sales setting, a salesperson that works in afast-paced environment may become highly stressed. However, if thesalesperson is provided the help of a support assistant, or a new tech-nology to automate some aspects of their job (e.g., automatic e-mailresponse, automated billing, etc.) the resource makes some aspects ofthe job easier, offsetting the high demand of other aspects of the job. Ina direct application of JD-R, we propose that the enhancement ofmorale through three job resources (sales capabilities training, firm'sproduct portfolio depth, and sales unit's cross-functional cooperation)will reduce the negative effect of market demands on sales forcemorale.

Finally, prior investigations utilizing the JD-R model have been

conducted at the individual employee level. However, recent con-ceptual work (Bakker & Demerouti, 2014, 2018), suggests that the JD-Rmodel can also be utilized at the organizational level. Applying the JD-R model at this level implies that leaders can redesign jobs to proac-tively provide job resources that offset current, or even anticipated jobdemands (Bakker & Demerouti, 2014). For example, increased marketdemands emanating from anticipating that the market is changing orthat new products will soon be introduced can be offset by im-plementing ongoing, non-intrusive training programs that prepare thesales force for the impending product launches. To the best of ourknowledge, the present research is the first study to test JD-R theory atthe sales force-level, thus making a novel contribution to the applica-tion of JD-R theory.

2.2. Hypotheses development

As shown in Fig. 1, we anticipate that two distinct demands stem-ming from the market – that is, customer purchase complexity (i.e.,longer, more demanding sales processes) and market dynamism (i.e.,changing environment) – will negatively impact sales force morale. Wealso expect that three organizational-level job resources (i.e., salescapabilities training, firm's product portfolio depth, and sales unit'scross-functional cooperation) will mitigate the negative impact of thetwo market demands on sales force morale. Finally, in keeping with thekey tenets of JD-R (e.g., Bakker & Demerouti, 2014), we anticipate thatenhanced levels of sales force morale will decrease sales force turnoverand increase sales force productivity. We next elaborate on the con-ceptual logic of our hypotheses.

2.3. Market demands and sales force morale

Customer purchase complexity. Consistent with Anderson, Chu, andWeitz (1987), we define customer purchase complexity as the extentto which sales force tasks entail working with a customer basewhose purchase decision-making processes involve long purchasetimes, high levels of information needs, and/or largely unfamiliarpurchase situations. The inclusion of purchase complexity is salient,given that modern sales forces face escalating and recurring com-plexity as they endeavor to provide product/service solutions thatserve client needs (Schmitz & Ganesan, 2014). In doing so, thecomplex nature of customer purchase decision processes often re-quires devotion of considerable time to navigating resources andcapabilities within their firms (e.g., sales teams, R&D, etc.), andexternal to their firms (e.g., agencies, outsourcers, etc.) (Plouffe,Bolander, Cote, & Hochstein, 2016; Schmitz & Ganesan, 2014).These requirements leave sales forces challenged to keep up with allthat is required of them to simply accomplish their core job task ofselling products. In essence, the demands of purchase complexityput the sales force in a difficult position with collective concernsthat client demands will go unmet because meeting them is nearlyimpossible.

Purchase complexity is viewed as a job demand because it is per-vasive, continual, and often at the edge of sales force capacities, whichcan lead to negative psychological consequences (Singh, Goolsby, &Rhoads, 1994). For sales forces that compete in industries with highpurchase complexity, job demands are high, as the sales force is taskedwith sustained activity concerning multiple customer related elementsto complete the sales process. Applying JD-R, we thus anticipate thatincreases in customer purchase complexity will reduce sales forcemorale. Hence,

H1. Customer purchase complexity has a negative relationship withsales force morale.

Market dynamism. Market dynamism refers to business environments

Fig. 2. Structure & levels of study variables.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

5

that they experience frequent change and shifts in focus (Joshi &Campbell, 2003). Market dynamism typically affects a firm in one ofthree main areas: customer, competitor and/or technology(Jaworski & Kohli, 1993). Jones, Brown, Zoltners, and Weitz (2005)provide an overview of market dynamism and its impact on theaforementioned areas in sales environments. With regard to custo-mers, dynamic change is evidenced by increased customer require-ments for varied product choices and expectations of the salespersonand selling firm. Regarding competition, markets are described ashypercompetitive, in part because product lifecycles are decreasing.With regard to technology, sales forces are required to adapt to ever-changing internal and external technology demands. In short,market dynamism is relevant because current business environ-ments are rapidly changing and less predictable than in the past(Nakata, Zhu, & Izberk-Bilgin, 2011).

The frequently changing characteristic of market dynamism createshigh levels of sales force uncertainty (Joshi & Campbell, 2003). Theuncertainty caused by market dynamism on various fronts (e.g., cus-tomers, competition, and/or technology), is consistent with the defi-nition of a job demand. As such, we consider market dynamism a jobdemand – that is, an organizational aspect of the job requiring sustainedphysical or psychological (cognitive or emotional) effort or skills(Bakker & Demerouti, 2007). Applying JD-R, we thus expect that in-creases in market dynamism will reduce sales force morale. Formallystated:

H2. Market dynamism has a negative relationship with sales forcemorale

2.3.1. The moderating effects of job resources

Sales capabilities training. Sales capabilities training is referred totraining interventions directed toward improving the sales force'stask-related knowledge, skills, and abilities (Kumar & Pansari,2016). This type of training is characterized as the implementationof specific sales skills and abilities through coaching and rewardsdesigned to better equip the sales force to address the demands ofevolving sales environments (Miao & Evans, 2012). Sales capabilitytraining can be accomplished in different ways. At the organiza-tional level, the goal of sales capability training is to provide in-struction and feedback that allows the sales force to better adapttheir collective sales strategies to new situations (Mallin & Pullins,2009). In essence, sales capability training is a management tooldesigned to equip sales forces with the tools needed to work smarter(Sujan, Weitz, & Kumar, 1994). This type of training is particularlyrelevant to the present study as the demands emanating from cus-tomer purchase complexity and market dynamism represent thecomplex situations and changing environments addressed by salescapability training.

Recent JD-R research has specifically proposed that organizationallevel job redesign be used to increase employee job resources and re-duce the negative, structural effects of organizational job demands(Bakker & Demerouti, 2014, pp. 17–18). Job redesign is characterizedby a structural intervention at the organizational level developed toinstitute positive changes to jobs, tasks, or the conditions of the job inan effort to collectively react to the negative effects of job demands(Bakker & Demerouti, 2014, p. 18). In essence, sales capability training(e.g., training directed at improving the sales force's task-relatedknowledge, skills, and abilities in proactive response to job demands) isby definition a job redesign mechanism. It is important to note that thepresent study is one of the first to empirically study the effects of jobredesign in a JD-R context. Thus, we use the conceptualization of jobredesign (Bakker & Demerouti, 2014, pp. 17–18) to hypothesize thatsales capabilities training is expected to perform as a buffer to the

negative effects of customer purchase complexity and market dyna-mism on sales force morale. Hence,

H3. Sales capabilities training weakens the negative effects of (a)customer purchase complexity and (b) market dynamism on sales forcemorale

Firm's product portfolio depth. A firm's product portfolio depth ischaracterized by firms carrying product lines comprising a largenumber of product variants (Sorescu, Chandy, & Prabhu, 2003). Theaim of product portfolio depth is to equip the sales force with moreproducts (i.e., resources) that better meet changing or complexcustomer demands (Johnson & Sohi, 2014). There are several rea-sons that greater product portfolio depth is considered positive forsales forces. First, the products available to the sales force providesolutions to meet customer needs and greater depth of product re-sources allows more flexibility in solutions in complex and dynamicsituations. Second, product portfolio depth can also have a limitingeffect on competitors by reducing the need for customers to look toother sources to meet current needs (Christensen, 2001). Third,greater product portfolio depth creates greater economies of scaleand thus more competitive offerings for the sales force (Sorescu,Chandy, & Prabhu, 2003). Finally, sales forces that offer greaterproduct depth are viewed as leaders and innovators, which differ-entiates them from other sales forces (Sorescu, Chandy, & Prabhu,2003). Given these positive aspects of product portfolio depth, weexpect that a deeper product portfolio offers more and better optionsto salespeople that allows them to cater to increasingly changingand complex customer needs, thereby buffering the negative effectsof customer purchase complexity and market dynamism on salesforce morale. Formally stated:

H4. Firm's product portfolio depth weakens the negative effects of (a)customer purchase complexity and (b) market dynamism on sales forcemorale

Sales unit's cross-functional cooperation. In the marketing literature,cross-functional cooperation has been studied in regard to the col-laboration of the sales force with marketing and/or research anddevelopment (e.g., Ernst, Hoyer, & Rubsaamen, 2010; Le Meunier-FitzHugh, Massey, & Piercy, 2011). Accordingly, we define cross-functional cooperation as the degree to which the sales unit co-operates with the marketing and research and development (R&D)units in the strategy making process (De Luca & Atuahene-Gima,2007). Cooperation is accomplished by “collaborative strategymaking” that includes shared goals, mutual understandings, sharedresources, shared vision, and rapport between the sales force andother functions (Le Meunier-FitzHugh, Massey, & Piercy, 2011).Cross-functional cooperation is desired because, without it, eachfunction pursues different goals and priorities creating conflict be-tween the functions (Ernst, Hoyer, & Rubsaamen, 2010). When co-operation is in place, positive outcomes are achieved (e.g., superiorvalue creation, and performance) and conflict is reduced (Guenzi &Troilo, 2007). These cross-functional cooperation outcomes re-present a job resource that is supportive of sales force efforts. Cross-functional cooperation is consistent with job resource elements be-cause cross-functional cooperation represents outcomes that arephysical, psychological, and organizational components of the jobthat improve the achievement of work goals, reduces job demands(and the associated costs), and/or stimulates learning (Demerouti,Bakker, Nachreiner, & Schaufeli, 2001). Utilizing the moderatingaspect of job resources found in JD-R theory (e.g., Bakker &Demerouti, 2007), cross-functional cooperation is expected to re-duce the negative effects of customer purchase complexity andmarket dynamism on sales force morale. Hence,

H5. Sales unit's cross-functional cooperation weakens the negativeeffect of (a) customer purchase complexity and (b) market dynamism

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

6

on sales force morale.

2.3.2. Key outcomes of sales force moraleAs previously mentioned, sales force morale is a “good attribute,” in

that higher levels of it are expected to lead to decreases in turnover, andincreases in productivity. Next, we define these two key outcomes andintroduce our final set of hypotheses.

Sales force turnover. Sales force turnover is the ratio of the total salesforce that has voluntarily left the firm at the end of a given timeperiod, compared to the total sales force at the beginning of the timeperiod. Given that the study of morale at the sales force level is new,we draw from prior work from the management domain that pro-vides evidence on the effect of morale on firm-level employeeturnover. Specifically, in a meta-analysis, for instance, morale wasfound to have a negative relationship with turnover intentions(Chang, Rosen, & Levy, 2009). In a different review of organiza-tional turnover antecedents (Hausknecht & Trevor, 2011), job sa-tisfaction and organizational commitment (i.e., dimensions ofmorale) were also found to be negatively related to firm-levelturnover. The conceptual logic linking firm-level morale to firm-level draws from prior work, which examines the influence of col-lective attitudes on firm-level outcomes (e.g., Chan, 1998). Specifi-cally, because morale reflects a collective sense of enjoyment withand attachment to the firm, salespeople develop shared expectationsabout strong ties and a greater sense of community, belonging,support, and cooperation (Hausknecht, Hiller, & Vance, 2008).Collectively, these mechanisms are viewed by salespeople as im-portant resources that salespeople desire to retain by sticking withthe current firm. Accordingly, we propose that sales force moralewill have a negative relationship with sales force turnover. Hence,

H6. Sales force morale has a negative relationship with sales forceturnover.

Sales force productivity. In keeping with recent work in management(Kim & Ployhart, 2014), we define sales force productivity as theefficiency of a firm's sales force to produce outputs and efficientlydeploy sales force human capital resources. The positive nature ofproductivity is inherent in the outcomes (e.g., increased revenues,profits, etc.) that are compared to efficiency (e.g., fewer employees,lower costs, etc.) to provide a measure of performance per worker.In essence, then, higher output from fewer salespeople reflectshigher productivity.

Prior work pinpoints that satisfied work units will set norms androutines for higher productivity, and enforce these norms and routineswithin the unit (e.g., Whitman, Van Rooy, & Viswesvaran, 2010). Inparticular, members of a unit that share positive feelings regarding theirfirm will tend to cooperate with mutual trust, resolve conflicts moreefficiently, and work better together to achieve the shared organiza-tional goals (Ostroff, 1992). We align our predictions with these find-ings and expect that in firms exhibiting higher levels of morale, as re-flected in higher feelings of satisfaction and commitment, the salesforce will perform at higher levels. Hence,

H7. Sales force morale has a positive relationship with sales forceproductivity.

3. Methods

3.1. Sample and data collection

Our research design involves a three-stage data collection that tookplace over two time periods (i.e., year1 and year2) and merged threedifferent sources: managers, salespeople, and secondary objective data.

Consequently, our research design alleviates concerns over commonmethod bias and endogeneity as a result of simultaneity, because itinvolves data from three different sources as well as temporal separa-tion of construct measurement over two time periods.

In the first stage, which took place in January–February of year1, wesolicited participation from 442 firms in the pharmaceutical, food &beverage, and electrical equipment industries in one European Unioncountry. We concentrated on industries (a) which were highly likely tobe intrigued by a study of salesperson morale due to their sales forcebeing an important component of firm strategy; (b) which should ex-hibit adequate variation on our key constructs; and (c) whose profes-sional associations were willing to grant us access to their rosters ofmember firms. One hundred and thirty-four (134) senior managersagreed to participate for a response rate of 30%. One of the authorsscheduled face-to-face meetings with the participating managers at thefirms' premises. The goal of these meetings was dual. First, we neededto establish rapport with participants to ensure that they will not onlypromote the study to their salespeople but also that they will participateagain in a brief survey at year2. We thus spend time explaining theobjectives and the stages of our research design to managers. Second,we needed to secure reliable measurement of study constructs giventhat responding to key questions required that managers contact ex-ecutives in other functional units (e.g., human resources) in order toobtain these figures (e.g., number of full-time salespeople that hadvoluntarily left the firm). Accordingly, we handed the survey off tomanagers and walked them through in order to eliminate any ambi-guities and to highlight the survey questions that would likely requireinput from other executives in the organization. Next, we asked man-agers to fill the survey out at a different time when the author was notpresent and to mail it back to the researchers. Managers provided re-sponses regarding market demands and job resources as well as otherkey metrics related to their B2B sales force1 (see Table 2). Given theinformants' high hierarchical position and long tenure in our study (seeAppendix A), responses are gauged to be reliable (Homburg, Klarmann,Reimann, & Schilke, 2012).

In the second stage, at the end of February of year1, managers dis-tributed a research packet to all members of their B2B sales force (3285in total). The packet contained a copy of the salesperson survey (whichcaptured responses related to sales force morale), a cover letter signedby the leading author that explained the purposes of the survey andassuring anonymity of responses, as well as a pre-stamped envelope,which salespeople used to mail their responses directly back to theUniversity, where the leading author was employed at the time of thestudy. To maximize response rates, we asked participating managers tosend two follow-up emails to their salespeople. We received 878 usefulsalesperson responses over a period of eight weeks for a 27% responserate across firms. On average, the salesperson response rate from eachparticipating firm was 52%, thus satisfying the criterion that has beenset forth in the literature for aggregating responses to the firm level(Wright et al., 2001). To assess any potential nonresponse bias, wecompared early (i.e., those responding before the first reminder) andlate (i.e., those responding after the second reminder) respondents forall study variables. No statistically significant differences were detected(p > .05).

In the third stage, in the beginning of December of year2, we com-municated with managers (that had participated in year1’s survey) andasked them to answer questions regarding sales force turnover duringthe period of the study. Because of firm/manager attrition or non-re-sponse, we were able to obtain answers from ninety-four (94) man-agers. Specifically, eight firms were merged/acquired; fifteen managersleft their firms during the two-year study period; and seventeen

1 Consistent with the objectives of our study, managers were explicitly instructed torespond to survey questions only for the B2B salespeople employed by their firms ratherthan for frontline employees, in general.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

7

managers didn't respond to the year2 survey. Next, we collected sec-ondary, objective data on year2 operating revenues for eighty-one (81)firms that participated in the year2 manager survey.2 It is this group of81 firms that constitutes the sample on which our hypothesis testinganalyses are based.

4. Measures

We drew measures for our constructs from existing scales, whereavailable (Table 2). To make sure there are no wording or ambiguityissues with the items, year1 surveys were pretested with a sample of 6managers and 17 salespeople before administering them to the targetpopulation. Surveys were administered in the local language after aback-translation procedure performed by one professional translatorand the leading author.

Manager-reported measures (year1). Customer purchase complexitywas measured with four reflective items designed to capture thethree dimensions of a complex purchase, as delineated in the workof Anderson, Chu, and Weitz (1987): long purchase cycles, highinformation needs, and high degree of newness of problem. Con-sistent with Homburg, Martin, and Wieseke (2012), market dyna-mism was operationalized as a formative index with three itemscovering the frequency of changes in competition, customer needs,

Table 2Study measures.

Manager-reported measures (at year1)

Customer purchase complexity: reflective measure; CR=0.76; AVE=0.45; 5-pointscale (strongly disagree, strongly agree)

The following statements refer to how customers in your market make purchase decisions.Please indicate the extent to which you agree with each of the following:

The purchase decision regarding goods/services in our market is made quickly(RS). (0.58)

Our customers need a lot of information before making a purchase decisionregarding goods/services in our market. (0.64)

Our customers consider the purchase decision regarding goods/services in ourmarket to be routine (RS). (0.78)

Our customers have routinized the purchase decision involving goods/services inour market so that it no longer requires a lot of attention (RS). (0.66)

Market dynamism: formative index; 5-point scale (very low frequency, very highfrequency)

Please indicate how frequently the following aspects change in your market:Goods/services offered by competition. (NA)Customer needs. (NA)Product technology. (NA)

Sales capabilities training: reflective measure; CR=0.85; AVE=0.54; 7-point scale(to no extent, to a great extent)

To what extent does your sales unit offer systematic training programs to your salespeoplein each of the following areas?

Interpersonal communication capabilities. (0.66)Understanding customer needs. (0.82)Understanding company policies/procedures. (0.72)Employing effective sales techniques. (0.89)Using sales technologies. (0.52)

Firm's product portfolio depth: manifest measure; 5-point scale (strongly disagree,strongly agree)

Please indicate the extent to which you agree with the following statement regarding theproduct lines your salespeople promote:

Our salespeople promote product lines comprising a large number of productvariants each. (NA)

Sales unit's cross-functional cooperation: formative index; 5-point scale (to no extent, toa great extent)

Please indicate the extent to which the sales unit in your company cooperates with each ofthe following units. In this company, the sales unit:

Fully cooperates with R&D in establishing goals and priorities for our strategies.(NA)

Fully cooperates with Marketing in establishing goals and priorities for ourstrategies. (NA)

Number of salespeople employed: manifest measureFor the current year, how many full-time salespeople are employed in your sales unit?

(NA)

Firm's customer portfolio size: manifest measureApproximately how many active customers does your sales unit currently serve? (NA)

Percentage of salary in sales compensation: manifest measureWhat is the percentage (%) of fixed salary in your salespeople's total annual

compensation? (NA)

Sales force team orientation: reflective measure; CR=0.76; AVE=0.52; 10-pointscale (to no extent, to a great extent)

The following statements refer to how your salespeople function as members of a team inyour sales unit. Please indicate the extent to which your salespeople:

Are willing to accept direction from their sales manager. (0.85)Cooperate as part of a sales team. (0.67)Accept the authority of their sales manager. (0.62)

Sales territory design effectiveness: reflective measure; CR=0.83; AVE=0.50; 7-pointscale (completely ineffective, completely effective)

The following statements refer to aspects of sales territory design in your sales unit. Pleaseindicate how effective your sales unit is on each of the following aspects:

The geographical size of our sales territories. (0.67)The number of calls made in our sales territories. (0.69)The amount of travel required in our sales territories. (0.74)The assignment of salespeople to our sales territories. (0.63)The overall design of our sales territories. (0.80)

Salesperson-reported measures (at year1)Sales Force Morale: second-order reflective measure; CR=0.86; AVE=0.68; 5-point

scale (strongly disagree, strongly agree)

Table 2 (continued)

Manager-reported measures (at year1)

The following statements refer to how you feel about your job and the company you'recurrently employed. Please indicate the extent to which you agree with each of thefollowing:

Affective Organizational Commitment (0.90)a

I talk up this company to my friends as a great organization to work for. (0.67)I find that my values and the company's values are very similar. (0.67)I am proud to tell others that I am part of this company. (0.79)This company really inspires the best in me in the way of job performance. (0.78)I am extremely glad I chose this company to work for over others I was consideringat the time I joined. (0.77)

I really care about the fate of this company. (0.54)For me, this is the best of all companies for which to work. (0.75)

Satisfaction with the job (0.71)a

My work gives a sense of accomplishment. (0.68)My job is exciting. (0.74)My work is satisfying. (0.84)I'm really doing something worthwhile in my job. (0.84)

Satisfaction with company policy & procedures (0.85)a

Management is progressive. (0.83)Top management really knows its job. (0.83)This company operates efficiently and smoothly. (0.84)Salespeople in company receive good support from the home office. (0.69)

Manager-reported measures (at year2)Sales force turnover: manifest measureAs of the present time, how many of your full-time salespeople, who wereemployed in your sales unit last year, have voluntarily left the sales unit?

Number of salespeople employed: manifest measureFor the current year, how many full-time salespeople are employed in your salesunit?

Secondary, objective measures (at year2)Sales force productivity: manifest measureFirm operating revenues (€)

a Second-order standardized loadings. RS= reverse scaled. NA=not ap-plicable for manifest/formative measures.

2 We collected operating revenues for 81 out of the 94 firms that participated in theyear2 survey because we could not obtain secondary financial data for 13 firms. This isbecause public disclosure requirements for private firms is less comprehensive in Europethan in the United States (see Homburg, Jensen, & Hahn, 2012 for a discussion on thesedifferences). Also, we made sure that secondary, objective data on year2’s operatingrevenues match the corresponding fiscal year for each of these 81 firms.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

8

and product technology. Sales capabilities training is a new re-flective measure developed on the basis of prior work (Kumar,Sunder, & Leone, 2014). It captures the extent to which firms sys-tematically offer training interventions directed toward improvingthe sales force's task-related knowledge, skills, and abilities in fivecritical areas. Firm's product portfolio depth is a single-item measuredeveloped on the basis of the work of Sorescu, Chandy, and Prabhu(2003) and is designed to capture the extent to which firms carryproduct lines comprising a large number of product variants. Wemeasured sales unit's cross-functional cooperation with two for-mative items, which are inspired and adapted from the work of DeLuca and Atuahene-Gima (2007) to fit the context of our study. Themeasure captures the extent to which the sales unit cooperates withmarketing and R&D; thus, because a sales unit may cooperate wellwith marketing but not necessarily with R&D, we treat the two itemsare formative. We measured the number of full-time salespeopleemployed in the sales unit at year1 with a single item; we use thismeasure to calculate sales force turnover at year2 (discussed sub-sequently). Finally, managers reported on the percentage (%) offixed salary in the sales force's total compensation and the size of thefirm's customer portfolio; we employ these two measures as cov-ariates in our hypotheses testing procedure in order to capture anyunobservable effects of type of compensation and sales opportunityon sales force turnover and productivity, respectively. Doing so isaligned with prior work, which has shown that sales force com-pensation or a larger sales territory are both related to sales forceoutcomes (e.g., Cravens, Ingram, LaForge, & Young, 1993).Salesperson-reported measures (year1). Consistent with prior work(e.g., Chang, Rosen, & Levy, 2009; Harrison, Newman, & Roth, 2006and Rosen, Levy, & Hall, 2006), sales force morale was measured asa second-order reflective construct with three first-order reflectivedimensions. Affective organizational commitment was measuredwith seven items drawn from Mowday, Porter, and Steers (1979).Satisfaction with the job and satisfaction with company policies/procedures were measured with four items each (Comer, Machleit, &Lagace, 1989).Manager-reported measures (year2). We operationalized sales forceturnover as the percentage of salespeople that have voluntarily leftthe firm at year2. Specifically, at year2, we asked managers to reportthe number of full-time salespeople – who were employed in theirsales unit at year1 – that voluntarily left the firm at year2. Next, weused this measure, together with the measure of the number of full-time salespeople employed in the sales unit at year1, to calculatesales force turnover at year2 as follows:

=

−

−

×

Sales Force Turnover (%)

#of full time salespeople employed at year1

that have voluntarily left the firm at year2#of full time salespeople employed at year1

100

YEAR2

(1)

In addition, we asked managers to report the number of full-timesalespeople employed in their sales unit at year2; we employ thismeasure to calculate sales force productivity.

Secondary, objective data (year2). In keeping with recent work onfirm productivity (Kim & Ployhart, 2014), we operationalize salesforce productivity as the ratio of firm operating revenues to totalnumber of full-time salespeople employed in the fiscal year fol-lowing survey data collection (year2). This productivity measure isessentially an indicator of total output to sales labor input and itthus captures the efficiency of the sales force to produce output (Kim& Ployhart, 2014). Furthermore, operating revenue is relevant in asales force context given that it refers to sales generated from acompany's day-to-day selling activities for which salespeople areprimarily responsible. Indeed, operating revenues has been em-ployed in prior sales research (e.g., Homburg, Jensen, & Krohmer,

2008). Secondary data on firm operating revenues come fromKantar TNS, a market research firm, which specializes in collectingfirm financial data in this European Union country. As mentionedpreviously, data on the number of full-time salespeople employed inthe sales unit at year2 come from the manager survey at year2. Wecalculate sales force productivity at year2 as follows:

=

−

Sales Force ProductivityFirm operating revenues (€)

#of full time salespeople employed

YEAR

YEAR

YEAR

2

2

2 (2)

4.1. Measure assessment

We assessed the factor structure and validity of our reflective con-structs through confirmatory factor analysis (CFA) in LISREL 8.80. Themodel for sales force morale in the full salesperson sample at year1(n=878) demonstrated good fit to the data: χ88

2= 337.35 (p < .01);RMSEA=0.06 (90% confidence interval for RMSEA: 0.05; 0.06);NFI= 0.98; NNFI= 0.98; CFI= 0.99; SRMR=0.04. Further, the re-sults support the second-order structure of morale, thus providingempirical evidence for the aggregation of morale's first order-factorsinto a second-order construct. Given these favorable results, consistentwith prior work in the area (Chang, Rosen, & Levy, 2009; Rosen, Levy,& Hall, 2006), we employ the second-order construct of sales forcemorale in our subsequent hypotheses testing. The average varianceextracted (AVE) was larger than 0.50 thus demonstrating convergentvalidity, whereas construct composite reliability was greater than the0.70 cut-off, thus offering support for its reliability (Table 2). All factorloadings were significant, with the lowest standardized loading equal to0.54 (t-value=14.45, p < .01). The results suggest that significantsystematic variance in the individual indicators can be attributed to theunderlying latent construct, thereby providing empirical support for thesecond-order reflective structure of our sales force morale scale.

Model specification for our full manager sample at year1 (n=134)included four multi-item reflective measures captured at year1 (cus-tomer purchase complexity, sales capabilities training, sales force teamorientation, and territory design effectiveness; see Table 2). The lattertwo measures are employed to examine any potential selection bias inour results (discussed subsequently). Results showed good fit: χ113

2=136.81 (p < .06); RMSEA=0.04 (90% confidence interval forRMSEA: 0.00; 0.06); NFI= 0.88; NNFI= 0.96; CFI= 0.97SRMR=0.06. Convergent validity was established because AVEs aregreater than or very close to the cut-off value of 0.50 (Table 2). Dis-criminant validity was indicated by the AVE for each construct beingsubstantially higher than its shared variance with any of the otherconstructs (see Tables 2 and 3). In addition, all factor loadings weresignificant, with the lowest standardized loading equal to 0.52 (t-value= 5.41 p < .01). Finally, all constructs show adequate levels ofreliability since their composite reliabilities fall above the cutoff of0.70.

4.2. Data aggregation & merging

According to our conceptual model, all constructs, including salesforce morale, reside at the firm-level (Figs. 1 and 2). Taking a firm-levelapproach in our analyses is consistent with prior employee morale re-search (e.g., Subramony, Krause, Norton, & Burns, 2008) and a directconsensus model in multilevel theory (Chan, 1998). Thus, in order toaggregate individual assessments of morale to the firm-level, we neededto assess whether sufficient agreement existed among sales forcemembers of each sales unit. To this end, we calculated the intraclasscorrelation (ICC[1]) for firms that returned at least 3 salesperson re-sponses (852 salespeople in 97 firms). The ICC[1] score was 0.23, ex-ceeding the recommended cutoff for justifying aggregation (Bliese,2000). We use this aggregated, firm-level measure of sales force morale

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

9

in subsequent hypothesis testing.Next, we created a firm-level file that matched together (a) the

salesperson-reported, aggregated sales force morale data at year1; (b)managers' responses on market demands and job resources at year1; (c)manager's reports of sales force turnover at year2; and (d) sales forceproductivity data at year2. For a firm to be included into this file, it hadto meet the following three criteria. First, a firm had to have returned atleast 3 salesperson surveys. As mentioned previously, we consider onlythe 97 firms that met this criterion (i.e., 852 salespeople in 97 firms).Second, a firm had to have participated to both year1 and year2 man-ager surveys. The number of firms that met this criterion was 94. Third,secondary, objective sales force productivity information had to bepublicly available. We were able to identify objective information for81 firms.

Against this backdrop, our final sample size employed in hypothesistesting is based on merged data from three sources (manager, sales-person, and secondary) and refers to 81 firms that met all three criteria:(a) firms that returned at least 3 salesperson surveys at year1 (528salespeople, in total); (b) firms whose managers participated at bothyear1 and year2 survey; and (c) firms for which secondary, objectiveinformation at year2 was available.

4.3. Model specification

Given the complexity of our model (containing multiple interactionterms and formative indexes) against sample size, we employ a partialleast squares (PLS) approach with SmartPLS 2.0 to test hypotheses. PLScan accommodate complex model specifications, such as nonlinear ef-fects, which are very difficult to be modeled with a covariance-basedapproach (Bagozzi & Yi, 2012). We employ a hierarchical model ap-proach. We first fit Model 1 to test main-effect hypotheses (H1, H2, H6

and H7). We then estimate Model 2 that includes direct effects ofmoderators. Finally, we fit Model 3 to test hypothesized interactions(H3a-b, H4a-b and H5a-b). Table 4 presents the results of our analysis.

5. Results

In regards to the hypothesized main effects (Model 1), results showthat customer purchase complexityyear1 exerts a negative effect on salesforce moraleyear1 (γ=−0.292, p < .01) in support of H1. Likewise,H2, which predicts a negative effect of market dynamismyear1 on sales

force moraleyear1 is confirmed (γ=−0.275, p < .01). As predicted inH6, sales force moraleyear1 negatively influences sales force turn-overyear2 (β=−0.200, p < .05). Also, consistent with our predictionsin H7, sales force moraleyear1 positively influences sales force pro-ductivityyear2 (β=0.231, p < .05). Although not formally hypothe-sized, Model 2 reports the main effects of the moderating variables.Specifically, we find that sales capabilities trainingyear1 is positivelyrelated to sales force moraleyear1 (γ=0.125, p < .05), sales unit'scross-functional cooperationyear1 is negatively related to sales forcemoraleyear1 (γ=−0.146, p < .05), whereas a firm's product portfoliodepthyear1 is not related to sales force moraleyear1 (γ=0.136, p > .10).

Turning our attention to the hypothesized interactions (Model 3),results show that, consistent with H3a and H3b, sales capabilities trai-ningyear1 weakens the negative effects of customer purchase complex-ityyear1 (γ=−0.216, p < .05) and market dynamismyear1

(γ=−0.241, p < .01) on sales force moraleyear1. In other words, wefind that customer purchase complexity's and market dynamism's ne-gative effects on morale are weakened when firms offer higher levels ofsales capabilities training.

However, in contrast to our predictions in H4a and H4b, the mod-erating effects of a firm's product portfolio depthyear1 on the relation-ship of customer purchase complexityyear1 (γ=0.235, p < .01) andmarket dynamismyear1 (γ=0.144, p < .05) with sales force mor-aleyear1 are positive. Thus, we find that the negative effects of customerpurchase complexity and market dynamism on morale are amplified,rather than weakened, when firms offer a deeper product portfolio.

Finally, H5a receives support since the moderating effect of salesunit's cross-functional cooperation on the relationship between cus-tomer purchase complexityyear1 and sales force moraleyear1 is negative(γ=−0.284, p < .01). This finding implies that when sales unitsenjoy higher levels of cooperation with marketing and R&D functions,the negative effect of purchase complexity is weakened. However, thehypothesized interaction between sales unit's cross-functional co-operationyear1 and market dynamismyear1 on sales force moraleyear1 isnot supported (γ=−0.131, p > .05); thus, H5b is not confirmed.

Beyond testing our hypothesized moderating effects for statisticalsignificance, we also wanted to examine their practical significance.Since Model 1 is nested within Model 3, we tested whether the twomodels differ significantly in terms of explained variance. Specifically,we compared the proportion of variance explained in sales force moraleby the main effects model (R1

2= 0.187) with that of the hypothesized

Table 3Descriptive statistics and construct intercorrelationsa,b.

Measures M SD 1 2 3 4 5 6 7 8 9

1. MORALEYEAR1c 3.94 0.322. CAPTRAINYEAR1

d 4.87 1.32 0.153. PPDEPTHYEAR1

d,f 3.36 1.32 0.10 0.35⁎⁎

4. CROSSFYEAR1d 2.83 0.85 −0.21 0.15 0.125. PCOMPLEXYEAR1

d 3.65 0.86 −0.34⁎⁎ −0.02 0.24⁎ 0.046. MDYNAMYEAR1

d 2.45 0.71 −0.32⁎⁎ 0.02 −0.08 0.45⁎⁎ 0.167. TURNOVER(%)YEAR2d,f 7.00 8.00 −0.18 −0.07 0.01 0.08 0.02 0.048. PRODUCIVITY(€)YEAR2e,f 337,699.96 373,830.84 0.18 0.22⁎ 0.02 0.13 0.04 −0.01 −0.229. SALARY(%)YEAR1d,f 70.98 19.81 −0.16 0.09 −0.02 −0.10 0.13 0.03 −0.13 0.30⁎⁎

10. FCPSYEAR1d,f 5723.83 8250.88 0.16 0.17 0.03 0.01 −0.08 −0.03 −0.02 −0.04 −0.29⁎⁎

PCMPLEX= customer purchase complexity; MORALE= sales force morale; MDYNAM=market dynamism; CAPTRAIN= sales capabilities training;PPDEPTH= firm's product portfolio depth; CROSSF= sales unit's cross-functional cooperation; TURNOVER= sales force turnover; SALARY=percentage of salaryin salesperson total compensation; FCPS= firm's customer portfolio size; PRODUCTIVITY= sales force productivity.

⁎ p < .05, two-tailed.⁎⁎ p < .01, two-tailed.a Pearson correlations based on standardized latent variable scores generated by the PLS algorithm.b Salesperson responses are aggregated to the firm level (n=81).c Salesperson-reported data.d Manager-reported data.e Secondary, objective data.f Manifest constructs measured with a single item.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

10

interactions model (R22= 0.438). The results reveal an effect size (f2)

of 0.45. The differences in explained variance between the two modelsrepresent strong effect sizes for the hypothesized moderating effects.

Finally, regarding the effects of covariates, we find that the per-centage of salary in total sales force compensation influences both salesforce turnover (γ=−0.172, p < .05) and sales force productivity(γ=0.339, p < .01). However, a firm's customer portfolio size was notfound to influence neither sales force turnover (γ=−0.041, p > .05)nor productivity (γ=0.027, p > .05).

5.1. Robustness checks

We assessed the extent to which our results are robust against (a)multicollinearity; (b) sample selection bias; and (c) higher-orderquadratic effects. First, we computed the inflation factors (VIFs) foreach predictor. The maximum VIF is 1.497, well below the thresholdvalue of 10 (Hair, Black, Babin, & Anderson, 2010), thus indicating nomulticollinearity issues.

Second, as mentioned previously, our analysis is based on a sub-sample of the firms that responded to our surveys and that met the threeinclusion criteria (see Data Aggregation & Merging section). This non-random selection process, however, might lead to biased parameterestimates if excluded firms are different than included firms. To explorethis issue, we adopted Heckman's (1979) selection model. Specifically,we estimated three probit selection models with the binary dependentvariable in each model capturing whether a firm met the specific cri-terion or not. Following Wooldridge (2009), we included the same setof variables used in hypotheses testing in each model, as well as twoadditional variables obtained from the manager survey at year1 aspredictors in this firm-level model (for details and results see Appendix

B). Specifically, we employed a three-item measure of sales force teamorientation (Cravens, Ingram, LaForge, & Young, 1993) and a five-itemmeasure of territory design effectiveness (Grant, Cravens, Low, &Moncrief, 2001) (see Table 2 for items). None of the probit models weresignificant (p > .10). Consequently, concerns over any effects of un-observed firm characteristics related to the selection process that mighthave biased our parameter estimates are alleviated.

Third, we empirically explored whether a firm's product portfoliodepthyear1 might be having a quadratic interaction (i.e., a non-linearmoderating effect) with customer purchase complexityyear1 and marketdynamismyear1 on sales force moraleyear1.3 Specifically, we estimatedModel 4 (see Table 4), which is identical to the hypothesized Model 3,but also contains the non-hypothesized quadratic term of a firm's pro-duct portfolio depthyear1, as well as the interactions between thequadratic term of a firm's product portfolio depthyear1 with both ourpredictors (i.e., customer purchase complexityyear1 and market dyna-mismyear1). The results indicate that neither the quadratic interaction ofa firm's product portfolio depthyear1 with purchase complexityyear1(γ=−0.050, p > .05) nor the quadratic interaction of a firm's productportfolio depthyear1 with market dynamismyear1 (γ=0.020, p > .05) issignificant thereby strengthening confidence in our hypothesized linearinteractions (i.e., H4a and H4b).

6. Discussion

Despite that the morale of an organization's sales force can have asubstantial impact on salesperson performance (Martin, 2015) as well

Table 4Results of structural equation analysesa,b,c.

Paths Main effects & covariates(Model 1)

Direct effects of moderators(Model 2)

Hypothesized interactions(Model 3)

Non-hypothesized quadratics(Model 4)

Endogenous variable: Sales Force MoraleYEAR1 R2= 18.7% R2=23.7% R2=43.8% R2= 44.2%PCOMPLEXYEAR1→MORALEYEAR1 −0.292⁎⁎ −0.329⁎⁎ −0.194⁎ −0.191⁎

MDYNAMYEAR1→MORALEYEAR1 −0.275⁎⁎ −0.195⁎⁎ −0.203⁎⁎ −0.182⁎

CAPTRAINYEAR1→MORALEYEAR1 – 0.125⁎ 0.108 0.106PPDEPTHYEAR1→MORALEYEAR1 – 0.136 0.211⁎⁎ 0.245⁎

CROSSFYEAR1→MORALEYEAR1 – −0.146⁎ −0.085 −0.088PPDEPTH2

YEAR1→MORALEYEAR1 – – – 0.049PCOMPLEXYEAR1×CAPTRAINYEAR1→MORALEYEAR1 – – −0.216⁎ −0.226⁎

PCOMPLEXYEAR1× PPDEPTHYEAR1→MORALEYEAR1 – – 0.235⁎⁎ 0.223⁎

PCOMPLEXYEAR1×CROSSFYEAR1→MORALEYEAR1 – – −0.284⁎⁎ −292⁎⁎

PCOMPLEXYEAR1× PPDEPTH2YEAR1→MORALEYEAR1 – – – −0.050

MDYNAMYEAR1×CAPTRAINYEAR1→MORALEYEAR1 – – −0.241⁎⁎ −0.234⁎

MDYNAMYEAR1×PPDEPTHYEAR1→MORALEYEAR1 – – 0.144⁎ 0.149⁎

MDYNAMYEAR1×CROSSFYEAR1→MORALEYEAR1 – – −0.131 −0.124MDYNAMYEAR1×PPDEPTH2

YEAR1→MORALEYEAR1 – – – 0.020

Endogenous variable: Sales Force Turnover YEAR2 R2= 5.9% R2=5.9% R2=5.9% R2= 5.9%MORALEYEAR1→ TURNOVERYEAR2 −0.200⁎ −0.200⁎ −0.200⁎ −0.200⁎

SALARYYEAR1→ TURNOVERYEAR2 −0.172⁎ −0.172⁎ −0.172⁎ −0.172⁎

FCPSYEAR1→ TURNOVERYEAR2 −0.041 −0.041 −0.041 −0.041

Endogenous variable: Sales Force ProductivityYEAR2 R2= 14.1% R2=14.1% R2=14.1% R2= 14.1%MORALEYEAR1→ PRODUCTIVITYYEAR2 0.231⁎ 0.231⁎ 0.231⁎ 0.231⁎

SALARYYEAR1→ PRODUCTIVITYYEAR2 0.339⁎⁎ 0.339⁎⁎ 0.339⁎⁎ 0.339⁎⁎

FCPSYEAR1→ PRODUCTIVITYYEAR2 0.027 0.027 0.027 0.027

Note: T-values are calculated from 1000 bootstrapped samples and are for one-tailed test for all directional hypotheses; Critical values: 1.65 (p < .05), 2.33(p < .01). N=81 firms. PCMPLEX= customer purchase complexity; MORALE= sales force morale; MDYNAM=market dynamism; CAPTRAIN= sales cap-abilities training; PPDEPTH= firm's product portfolio depth; CROSSF= sales unit's cross-functional cooperation; TURNOVER= sales force turnover;SALARY=percentage of salary in salesperson total compensation; FCPS= firm's customer portfolio size; PRODUCTIVITY= sales force productivity.

⁎ p < .05.⁎⁎ p < .01.a Entries within each cell correspond to standardized path estimates.b Explained variance (R2) indicates good explanatory power of the proposed model.c Stone-Geisser (Q2) cross-validated redundancy values, yielded from a blindfolding procedure (d=7), are 0.082, 0.054, and 0.079 for sales force morale, sales

force turnover, and sales force productivity respectively; these values show that the model exhibits good predictive relevance.

3 We would like to thank an anonymous Reviewer for suggesting this set of analysis.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

11

as the high level of interest from managers in improving morale, threeimportant research questions/gaps remain: (1) What market demandsnegatively impact sales force morale? (2) What resources can an or-ganization leverage to buffer the negative effects market demands haveon sales force morale? and (3) What impact does sales force moralehave on key organizational outcome measures? Our study addressesthese questions/gaps and contributes to theory and managerial practicein the following ways:

6.1. Theoretical implications

Overall, our research contributes to the JD-R literature by applyingJD-R at to a study of organizational level topics. The current researchspecifically answers a very recent call of Bakker and Demerouti (2018)stated as “we suggest in JD-R theory that well-being and performanceare the outcomes of factors at the individual (job function) level butalso at the team or even the organizational level. Several studies haveprovided evidence for such a claim. However, the empirical evidence isstill scarce and scattered. We need more empirical evidence regardingwhether a factor at the organizational level consistently acts as buffer orexacerbates…well-being” (p. 8). Interestingly, our findings in a salessetting largely support the foundational tenets of JD-R. Yet, as indicatedby Bakker and Demerouti (2018) we discover that, indeed, resourcescan buffer and exacerbate organizational job demands (discussed sub-sequently in more detail). This finding represents a difference in howjob resources have operated in previous research at the individual-level.In addition, our research also adds in more specific ways, described inthe following sections.

First, despite the unique and direct impact sales forces have on or-ganizations' performance and the impact morale has on salespersonperformance (Martin, 2015), it is surprising that prior theoretical workhas not examined sales force morale and its antecedents. Our study fillsthis void by conceptualizing sales force morale as an organizational-level phenomenon and testing how market factors affect sales forcemorale. Utilizing a unique dataset comprising three different datasources collected in two time periods, we provide theoretical and em-pirical support that sales force morale is a higher-order construct that,though originating at the individual salesperson level, manifests at theorganizational level, consistent with the logic of compositional multi-level models (Chan, 1998). Furthermore, taking a JD-R perspective, weselect and evaluate how customer purchase complexity and market dy-namism can hurt sales force morale, as it is often contemplated, but notempirically tested, in prior work (e.g., Zoltners, Sinha, & Lorimer,2008). We show that both conditions can indeed lower sales forcemorale and thus lead to less positive feelings toward the organizationand the work environment. As such, highly complex and dynamicmarkets can actually hurt key sales force outcomes by lowering salesforce morale.

Second, we contribute to knowledge on how organizational levelresources can buffer the negative effects of markets demands on salesforce morale. Specifically, we address how and to what extent (a) salescapabilities training; (b) firm's product portfolio depth; and (c) salesunit's cross-functional cooperation can mitigate the negative effects ofdemanding conditions on sales force morale. Our results indicate thatoffering training that improves sales force capabilities acts as a buf-fering job resource in that salespeople are better able to cope with thedemands of a complex and dynamic market and thus mitigate thelowering of their morale. This finding is notable and suggests that of-fering training interventions equips the sales force with the necessaryknowledge, skills, and abilities that allow them to find new and novelways for addressing the challenges in their work. For example, whenconfronted with constantly changing customer preferences, capablesalespeople can utilize sales technology to quickly configure new so-lutions that can satisfy customers' emergent needs.

Our results also reveal that a cooperative relationship between thesales unit and functions such as marketing and R&D can mitigate the

negative influence of customer purchase complexity on sales forcemorale. We feel that this buffering effect is the result of sales workingharmonically with functions that play a critical role in the developmentof a deep understanding of customer needs or in the deployment ofcustomer solutions. This means that cross-functional competition fortime and resources is avoided, thereby allowing salespeople to mini-mize the personal costs associated with searching for solutions orgaining a deep understanding of customer needs. We also find that,though in the hypothesized direction (i.e., negative), the interactionbetween cross-functional cooperation and market dynamism is notsignificant. Therefore, while cross-functional cooperation between salesand other functions is instrumental for dealing with complexity in thepurchasing process, the same is not true when dealing with highlydynamic markets. We feel that this organizational-level finding illus-trates that change may be faster than firms can respond at a higher (i.e.,organizational) level. We suspect that this interaction would producedifferent results at the team- or individual- level, as the levels closer tothe market can be more nimble in response to change. Alternatively, itcould be the case that the level of uncertainty and uncontrollabilityimplied by market dynamism requires more drastic resources than justcross-functional cooperation to be put in place in order for the salesforce to cope with this type of environment. This conjecture is in linewith recent work (Rouziès & Hulland, 2014) suggesting that coopera-tion between sales and marketing in the presence of powerful customersmay actually “stifle innovation and inhibit rapid responses to changingmarket conditions because managers cannot access knowledge neces-sary to identify market opportunities” (p. 522) Overall, this finding,which deserves further empirical testing, speaks to the fact that oneresource does not fit all environmental contexts and that there may beother resources to be employed depending on the forces of the en-vironment that are more salient for a given company (i.e., customerpurchase complexity or market dynamism).

Our results regarding a firm's product portfolio depth are intriguing.Specifically, contrary to our predictions, we find that organization-levelresources intended to assist the sales force achieve their goals can ac-tually have unintended negative consequences. Specifically, the find-ings show that a deeper product portfolio magnifies the negative effectsof a demanding selling context on sales force morale. While having abroad product offering can provide salespeople with more ways to meetcustomer needs, the increased requirements for learning about orbuilding expertise with a prolific portfolio may be a daunting task,especially for salespeople that sell sophisticated products with highlearning curves. As such, a deeper product portfolio may actuallyburden the sales force with higher customer expectations while di-minishing their ability to be experts (Jones, Brown, Zoltners, & Weitz,2005). Likewise, a deeper product portfolio implies a greater degree ofheterogeneity in the attributes of products marketed, thereby increasingthe complexity of product space that salespeople face (Lenk, DeSarbo,Green, & Young, 1996). It is thus possible that product complexity in-fluences the search, evaluation, and opportunity costs associated withchoosing the right configuration of products that best meet changing orcomplex customer needs (Barroso & Giarratana, 2013). Consequently,the increased cognitive effort, and possible information overload, dueto greater product heterogeneity (Fernhaber & Patel, 2012) mightamplify the demands already stemming from a turbulent environment,thus further lowering sales force morale. These results draw attention tothe need for a better understanding of how job resources that arecommonly assumed to offer an advantage, may, in fact, produce ne-gative unintended consequences on sales force morale as these re-sources interact with forces of the external environment. We believethat these results offer novel insights and advance the application of JD-R theory in sales research by highlighting a less salient view in theextant literature – that is, adding more complexity on job resources canresult in poorly implemented resources (Bakker & Demerouti, 2007).These results also raise an interesting question regarding what firms cando to strike a balance between the need to offer a deeper product

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

12

portfolio to strategically fend off competitive threats in turbulentmarkets and the ability of salespeople to deal with this increasingcomplexity. We address this intriguing question in the “Limitations andFuture Research Directions” subsection.

Third, we contribute to the literature by showing that sales forcemorale significantly affects objective sales force productivity andturnover rates, which are captured a year after the administration of thesurvey measuring sales force morale. Managers are increasingly facedwith highly demanding environments characterized, in part, by thecomplexity of customers' buying decision process and dramaticallychanging conditions in their markets (Ledingham, Kovac, & Simon,2006). These conditions make it increasingly difficult for managers toimplement strategies that will prevent salesperson turnover or increasesales productivity (Sunder, Kumar, Goreczny, & Maurer, 2017). Whilethe extant literature offers important insights on how individual-levelnegative stressors, such as role ambiguity or burnout, can have detri-mental effects for salesperson performance and turnover intentions(e.g., Behrman & Perreault, 1984; Singh, Goolsby, & Rhoads, 1994), wecurrently know very little about the role sales force morale has in in-fluencing organizational-level outcomes. We bridge this gap and showthat modeling sales force morale is capable of explaining 14.1% of thevariance in sales force productivity and 5.9% of the variance in salesforce turnover. Because our research design involves data from differentsources captured with a temporal separation, our results cannot beexplained by common method bias or endogeneity as a result of si-multaneity. Also, because our model specification accounts for anypotential selection bias our results cannot be explained by self-selection– based endogeneity.

Finally, in contrast to prior work in marketing, management, andpsychology, our research addresses job demands and job resources atthe organizational-level, as opposed to the individual level, by ex-tending JD-R theory to organization level phenomenon. Specifically, weshow that organizational level job resources and demands are pre-dictive of collective attitudes (i.e. sales force morale) and outcomes (i.e.sales force productivity and turnover) at the organizational-level. Thesefindings provide empirical support for yet unconfirmed theoreticaldiscussions in the management literature (see Bakker & Demerouti,2014). Thus, we offer a novel insight regarding the applicability of theJD-R perspective at the organizational-level that is useful not only tomarketing but also to management scholars.

6.2. Managerial implications

Our study offers crucial insights for sales executives. First, we pro-vide evidence that sales force morale can positively affect key outcomesin that it helps increase sales force productivity while lowering volun-tary turnover rates. Specifically, we find that an increase of morale byone point on a 5-point scale improves sales force productivity by€226,834 of operating revenues per salesperson while it lowers turn-over rate by 5%.4 These findings are particularly relevant for firmsacross industries that face challenges with high levels of turnover, suchas software and medical device (Barton & Davis, 2016), or seek toimprove sales force productivity. The fact that we use objective data aswell as a time lag between measurements should lend confidence tomanagers when interpreting our findings.

Given these encouraging results, a second managerial implication ofour study stems from the finding that sales force morale is a constructthat pertains to the sales force's collective attitudes toward the orga-nization. In particular, our study informs managers on the steps theymight want to consider while measuring and monitoring morale. Inparticular, managers may want to focus on the dimensions of affective

organizational commitment, satisfaction with the job, and satisfactionwith company policy and procedures when designing initiatives aimedat measuring morale. These dimensions, which can conveniently bemeasured with the items shown in Table 2, do not take up much ad-ditional space in an existing company survey, thus making it easy forfirms to include them as part of their standard procedures for mon-itoring sales force behaviors and attitudes. In addition, these items weretested across a variety of firms in three industries that should makethem applicable to a number of different contexts.

Third, regarding the role of market demands, we find that bothcustomer purchase complexity and market dynamism lower sales forcemorale. Although managers have little, if any, influence on dimensionsof the external environment (e.g., managers cannot completely purgecomplexity or dynamism that stem from technological advancements,continuously evolving customer needs, and competitive rivalry), theycan, however, manage how salespeople deal with complexity and dy-namism. Our research suggests that job resources, pertinent to the in-ternal organizational environment, and thus under the control of thefirm, can mitigate the negative effects of either purchase complexity ormarket dynamism. Specifically, training the sales force with a goal ofimproving their sales capabilities can weaken the negative impact ofpurchase complexity and market dynamism on morale. For instance, ifa sales job entails high complexity and dynamism (e.g., working withcustomers that need a lot of information before making a purchasecommitment or when customer needs change frequently), salespeopleneed to be trained in areas such as understanding customer needs orcommunicating effectively with customers to acquire the capabilitiesneeded for mitigating the influence of complexity and dynamism.

In addition, salespeople working within a sales unit that enjoys aharmonious cooperation with R&D and marketing in establishing goalsand priorities for firm strategies seem to be better able to deal withcomplex purchasing decisions. It is possible that in such purchasingsituations salespeople are required to spend more time with the cus-tomer or even negotiate the customization of solutions internally withother functional units. As such, high levels of cross-functional co-operation with units such as R&D and marketing can serve a critical rolein dealing with complexity. However, this beneficial effect of cross-functional cooperation was not supported in the case of market dyna-mism. It thus appears that managers may want to ensure that their unithas built a good relationship with other functions, especially when highlevels of purchasing complexity characterize the environment.

Finally, our study shows that a firm's product portfolio depth canhave unintended negative consequences for the sales force. We findwhen sales forces are tasked with promoting a deep product portfoliothe negative effects that purchase complexity and market dynamismhave on sales force morale are amplified. Perhaps this happens becausea deep product portfolio implies product lines that comprise a largenumber of variants. This situation, however, entails that salespeopleneed to understand a lot of different products and their technical spe-cifications. At the same time, because different products might bemanaged by product managers that can often have different or evencompeting goals, salespeople may experience a depletion of personalresources while trying to strike a balance between competing productlines. This situation, in turn, can make salespeople feel that they are lesscapable to cope with a complex or dynamic market, and thus theirmorale will be further reduced. However, research on product portfoliomanagement has also shown that benefits of a deeper product portfoliomay be contingent on other factors such as whether the firm usesproduct specialists or generalists in their sales force (Zoltners, Sinha, &Lorimer, 2006). Specifically, firms that employ product specialist salesforces can benefit from their expertise in a specific product line and canpossibly benefit from lower salesperson turnover as specialists may bemore apt to remain in a niche area that comprises fewer selling com-panies.5 Managers should acknowledge such contingencies uncoveredby our research and try to manage their occurrence, perhaps by pro-viding additional support or coaching to their sales force.

4 These results are based on the unstandardized coefficients produced by two simpleregression model specifications where (a) sales force productivity and (b) sales forceturnover are regressed onto sales force morale, respectively.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

13

7. Limitations and future research

Our study is subject to several limitations. First, the job demands wefocus on in our study are by no means exhaustive of the potential de-mands that salespeople may face in their work. Specifically, one mightenvision exploring environmental forces that go above and beyond theinfluence of customers, competitors, and technology. For instance, po-litical, regulatory or sustainability forces are all having impacts on firmsand salespeople.

Second, our focus here is on demands stemming from the externalenvironment. However, demands may also refer to internal elementssuch as lack of leadership or coworker support as well as or a workclimate that emphasizes high levels of workload, thus contributing tonegative feelings about the job (Bakker & Schaufeli, 2000). In addition,because role ambiguity and role conflict are common to sales jobs, thesesalesperson-level demands should be investigated in future, moregranular, individual-level research of morale.6

Third, average variance extracted of our measure of customer pur-chase complexity falls below the recommended value of 0.50. Althoughsupporting evidence of the discriminant validity of this measure comesfrom the fact that it satisfies the stringent test recommended by Fornelland Larcker (1981) in that its AVE exceeds the highest squared corre-lation with any other construct (Table 3), additional work is needed infuture research to refine this construct.

Fourth, regarding job resources, we focus on organizational aspectsof the work environment. However, recent research suggests thatsalespeople may draw on resources such as the social capital they havedeveloped with people residing outside the organization such as ex-ternal business partners (Plouffe, Bolander, Cote, & Hochstein, 2016) orthe social support from one's family (Bakker, Demerouti, & Dollard,2008) which can be functional in achieving sales goals or in helpingreduce job demands. Thus, delving into the full array of sales resourcesand how they might affect morale is a fruitful avenue for future re-search.

Fifth, related to job resources, although our measure of “sales cap-abilities training” captures the systematic training that is offered tosalespeople, it does not explicitly captures the training interventions

that have been offered in the past. As such, future studies need tocontrol for the offering of past training in order to isolate the effect ofcurrent offerings in a more explicit way.7 Likewise, our measure of afirm's product portfolio depth comprises a single item. Although priorresearch shows that constructs with concrete singular objects and at-tributes that managers confront in their everyday work can be mea-sured with single-item scales and that predictive validity of single itemmeasures is as good as that of multiple-item measures (Bergkvist andRossiter, 2007), future investigators might want to further refine thisconstruct.

Sixth, the finding that product portfolio depth amplifies, rather thanalleviates, the negative effects of market demands on sales force moralesuggests exciting avenues for future research. On one hand, if due tocompetitive pressures firms must have a complex product portfolio,what can then they do to boost sales force morale? An interesting areafor research would be to examine the role of sales forces' dynamiccapabilities such as absorptive capacity or ambidexterity for helpingdeal with increasing levels of complexity in the market (Fernhaber &Patel, 2012). On the other hand, future investigators need to ascertainwhether a firm's product portfolio depth functions more as a job de-mand rather than a job resource. Doing so would perhaps requireconducting qualitative research that would help glean insights fromsalespeople regarding the role of this construct.

Finally, our selection of industries and firms was guided by practicalconsiderations and difficulties (i.e., collecting data from busy managersover two different points in time as well as from their salespeople). Assuch, we concentrated on three industries, which traditionally employ alarge number of salespeople and are thus interested in learning aboutways to improve sales force morale. This process however might havebiased our results in the sense that only firms with a large sales forceand a higher interest in sales force topics are included in our study.Accordingly, future research needs to be conducted in other industries(e.g., energy).

Acknowledgment: The first author would like to thank his threeresearch assistants for their help during the data collection stage of thisstudy. The authors would like to thank the three anonymous reviewersfor their helpful feedback.

Appendix A. Sample characteristics

Managers' sample Percentage (%) Salespeople's sample Percentage (%)

Firms' industryPharmaceuticalsFood & beveragesElectrical equipment

34.327.638.1

GenderFemaleMale

19.780.3

Firms' number of employees<500501+

86.913.1

EducationHigh SchoolCollegeUniversityGraduate School

27.421.743.87.0

Firms' annual sales (€)<5 million5–100 millions101–999 millions> 1 billion

15.959.523.01.6

Age<30 years30–39 years40–49 years50–59 years> 60 years

25.054.413.36.70.6

5 We would like to thank an anonymous Reviewer for contributing this insight.

6 We would like to thank an anonymous Reviewer for contributing this insight.7 We would like to thank an anonymous Reviewer for contributing this insight.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

14

Informant titleSales Director/VP of Sales/Sales ManagerGeneral Manager/Managing DirectorCommercial DirectorOwner/CEO/President

80.67.510.41.5

Total sales experience<1 year1–2 years3–6 years7–10 years11–20 years> 21 years

6.411.825.922.023.311.6

Informant genderMaleFemale

93.36.7

Organizational Tenure<1 year1–2 years3–6 years7–10 years11–20 years> 21 years

0.539.726.816.411.15.4

Informant organizational tenure<1 year1–5 years6–10 years11–20 years> 20 years

3.841.528.517.78.5

Appendix B. Assessment of selection bias-based endogeneity

A. For the criterion of including firms based on whether they returned at least 3 salesperson responses, we specify the Heckman's selection model asfollows:

= + × + × + × + × + × + × + ×

+

β β PCOMPLEX β MDYNAM β CAPTRAIN β PPDEPTH β CROSSF β TEAM β TERRITORY

ε

AVAIL

(W2.1)0 1 2 3 4 5 6 7

where:AVAIL=binary variable indicating whether at least three salespeople's responses are available, PCOMPLEX= customer purchase complexity,

MDYNAM=market dynamism, CAPTRAIN= sales capabilities training, PPDEPTH=product portfolio depth, CROSSF= sales unit's cross-func-tional cooperation, TEAM= sales force team orientation, TERRITORY= territory design effectiveness.

Results show that no parameter estimate is significant at p < .05 (likelihood-ratio χ72= 4.92, p > .10).

B. For the criterion of including firms based on whether manager responses were available at both year1 and year2 surveys, we specify theHeckman's selection model as follows:

= + × + × + × + × + × +β β MORALE β SALARY β CUSTOM β TEAM β TERRITORY εAVAIL (W2.2)0 1 2 3 4 5

where:AVAIL=binary variable indicating whether manager responses were available at both year1 and year2 surveys, MORALE= sales force morale,

SALARY=percentage of fixed salary in salesperson total compensation, CUSTOM=number of customers per salesperson, TEAM= sales force teamorientation, TERRITORY= territory design effectiveness.

Results show that no parameter estimate is significant at p < .05 (likelihood-ratio χ52= 1.45, p > .10).

C. For the criterion of including firms based on whether secondary, objective sales force productivity data at year2 were available, we specify theHeckman's selection model as follows:

= + × + × + × + × + × +β β MORALE β SALARY β CUSTOM β TEAM β TERRITORY εAVAIL (W2.3)0 1 2 3 4 5

where:AVAIL=binary variable indicating whether manager responses were available at both year1 and year2 surveys, MORALE= sales force morale,

SALARY=percentage of fixed salary in salesperson total compensation, CUSTOM=number of customers per salesperson, TEAM= sales force teamorientation, TERRITORY= territory design effectiveness.

Results show that only territory design effectiveness (β5=-0.484, p < .01) is significantly related to the availability of secondary, objective salesforce productivity data at year2 is significant; the rest of the parameters as well as the model (likelihood-ratio χ5

2= 7.97, p > .10) are notsignificant.

References

Anderson, E., Chu, W., & Weitz, B. (1987). Industrial purchasing: An empirical explora-tion of the buyclass framework. Journal of Marketing, 51(3), 71–86.

Baehr, M. E., & Renck, R. (1958). The definition and measurement of employee morale.Administrative Science Quarterly, 3(2), 157–184.

Bagozzi, R. P., & Yi, Y. (2012). Specification, evaluation, and interpretation of structuralequation models. Journal of the Academy of Marketing Science, 40(1), 8–34.

Bakker, A. B., & Demerouti, E. (2007). The job demands-resources model: State of the art.Journal of Managerial Psychology, 22(3), 309–328.

Bakker, A. B., & Demerouti, E. (2014). Job demands–resources theory. Wellbeing. Vol.3John Wiley & Sons, Ltd.

Bakker, A. B., & Demerouti, E. (2018). Multiple levels in job demands-resources theory:Ibmplications for employee well-being and performance. Handbook of well-beingSaltLake City, UT: DEF Publishers.

Bakker, A. B., Demerouti, E., & Dollard, M. F. (2008). How job demands affect partners'experience of exhaustion: Integrating work-family conflict & crossover theory.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

15

Journal of Applied Psychology, 93(4), 901–911.Bakker, A. B., Demerouti, E., & Verbeke, W. (2004). Using the job demands-resources

model to predict burnout and performance. Human Resource Management, 43(1),83–104.

Bakker, A. B., & Schaufeli, W. B. (2000). Burnout contagion processes among teachers.Journal of Applied Social Psychology, 30(11), 2289–2308.

Barney, J. B. (1986). Organizational culture: Can it be a source of sustained competitiveadvantage? Academy of Management Review, 11(3), 656–665.

Barroso, A., & Giarratana, M. S. (2013). Product proliferation strategies and firm per-formance: The moderating role of product space complexity. Strategic ManagementJournal, 34(12), 1435–1452.

Barton, S., & Davis, M. (2016). Turnover rates for sales employees reach a five-year high.accessed online March 28, 2018 at https://radford.aon.com/insights/articles/2016/Turnover-Rates-for-Sales-Employees-Reach-a-Five-Year-High.

Behrman, D. N., & Perreault, W. D. (1984). A role stress model of the performance andsatisfaction of industrial salespersons. Journal of Marketing, 48(4), 9–21.

Bergkvist, L., & Rossiter, J. R. (2007). The predictive validity of multiple-item versussingle-item measures of the same constructs. Journal of Marketing Research, 44(2),175–184.

Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability:Implications for data aggregation and analysis. In K. J. Klein, & S. W. J. Kozlowski(Eds.). Multilevel theory, research, and methods in organizations: Foundations, extensions,and new directions (pp. 349–381). San Francisco, CA: Jossey-Bass.

Britt, T. W., Dickinson, J. M., & Adler, A. B. (2006). Morale during military operations: Apositive psychology approach. In T. W. Britt, & C. A. Castro (Vol. Eds.), Military life:The psychology of serving in peace and combat: Military performance. Vol. 1. Westport,CT: Praeger Security International.

Campbell, D. T., & Tyler, B. B. (1957). The construct validity of work-group moralemeasures. Journal of Applied Psychology, 41(2), 91.

Chan, D. (1998). Functional relations among constructs in the same content domain atdifferent levels of analysis: A typology of composition models. Journal of AppliedPsychology, 83(2), 234.

Chang, C. H., Rosen, C. C., & Levy, P. E. (2009). The relationship between perceptions oforganizational politics and employee attitudes, strain, and behavior: A meta-analyticexamination. Academy of Management Journal, 52(4), 779–801.

Christensen, C. M. (2001). The past and future of competitive advantage. MIT SloanManagement Review, 42(2), 105–109.

Churchill, G. A., Ford, N. M., & Walker, O. C. (1976). Organizational climate and jobsatisfaction in the salesforce. Journal of Marketing Research, 13(4), 323–332.

Comer, J. M., Machleit, K. A., & Lagace, R. R. (1989). Psychometric assessment of areduced version of Indsales. Journal of Business Research, 18(4), 291–302.

Cotham, J. C., III (1968). Job attitudes and sales performance of major appliancesalesmen. Journal of Marketing Research, 5(4), 370–375.

Cravens, D. W., Ingram, T. N., LaForge, R. W., & Young, C. E. (1993). Behavior-based andoutcome-based salesforce control systems. Journal of Marketing, 57(4), 47–59.

Cron, W. L. (2017). Macro sales force research. Journal of Personal Selling & SalesManagement, 37(3), 188–197.

De Luca, L. M., & Atuahene-Gima, K. (2007). Market knowledge dimensions and cross-functional collaboration: Examining the different routes to product innovation per-formance. Journal of Marketing, 71(1), 95–112.

Deeter-Schmelz, D. R., Goebel, D. J., & Kennedy, K. N. (2008). What are the character-istics of an effective sales manager? An exploratory study comparing salesperson andsales manager perspectives. Journal of Personal Selling & Sales Management, 28(1),7–20.

Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands-resources model of burnout. Journal of Applied Psychology, 86(3), 499–512.

Deshpande, R., & Webster, F. E., Jr. (1989). Organizational culture and marketing:Defining the research agenda. Journal of Marketing, 53(1), 3–15.

Dickson, M. W., Resick, C. J., & Hanges, P. J. (2006). When organizational climate isunambiguous, it is also strong. Journal of Applied Psychology, 91(2), 351.

Dickson, P. R., Farris, P. W., & Verbeke, W. J. (2001). Dynamic strategic thinking. Journalof the Academy of Marketing Science, 29(3), 216–237.

Dubinsky, A. J., Howell, R. D., Ingram, T. N., & Bellenger, D. N. (1986). Salesforce so-cialization. Journal of Marketing, 50(4), 192–207.

Ernst, H., Hoyer, W. D., & Rubsaamen, C. (2010). Sales, marketing, and research-and-development cooperation across new product development stages: Implications forsuccess. Journal of Marketing, 74(5), 80–92.

Fernhaber, S. A., & Patel, P. C. (2012). How do young firms manage product portfoliocomplexity? The role of absorptive capacity and ambidexterity. Strategic ManagementJournal, 33(13), 1516–1539.

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with un-observable variables and measurement error. Journal of Marketing Research, 18(1),39–50.

Grant, K., Cravens, D. W., Low, G. S., & Moncrief, W. C. (2001). The role of satisfactionwith territory design on the motivation, attitudes, and work outcomes of salespeople.Journal of the Academy of Marketing Science, 29(2), 165–178.

Guenzi, P., & Troilo, G. (2007). The joint contribution of marketing and sales to thecreation of superior customer value. Journal of Business Research, 60(2), 98–107.

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis:A global perspective. Upper Saddle River, NJ: Pearson.

Harrison, D. A., Newman, D. A., & Roth, P. L. (2006). How important are job attitudes?Meta-analytic comparisons of integrative behavioral outcomes and time sequences.Academy of Management Journal, 49(2), 305–325.

Hartmann, N. N., & Rutherford, B. N. (2015). Psychological contract Breach's antecedentsand outcomes in salespeople: The roles of psychological climate, job attitudes, andturnover intention. Industrial Marketing Management, 51, 158–170.

Hausknecht, J. P., Hiller, N. J., & Vance, R. J. (2008). Work-unit absenteeism: Effects ofsatisfaction, commitment, labor market conditions, and time. Academy of ManagementJournal, 51(6), 1223–1245.

Hausknecht, J. P., & Trevor, C. O. (2011). Collective turnover at the group, unit, andorganizational levels: Evidence, issues, and implications. Journal of Management,37(1), 352–388.

Heckman, J. J. (1979). Sample selection bias as a specification error. Econometrica,47(January), 153–161.

Homburg, C., Jensen, O., & Hahn, A. (2012). How to organize pricing? Vertical delegationand horizontal dispersion of pricing authority. Journal of Marketing, 76(September),49–69.

Homburg, C., Jensen, O., & Krohmer, H. (2008). Configurations of marketing and sales: Ataxonomy. Journal of Marketing, 72(March), 133–154.

Homburg, C., Klarmann, M., Reimann, M., & Schilke, O. (2012). What drives key in-formant accuracy? Journal of Marketing Research, 49(4), 594–608.

Homburg, C., Martin, A., & Wieseke, J. (2012). Marketing performance measurementsystems: Does comprehensiveness really improve performance? Journal of Marketing,76(3), 56–77.

Ingram, T. N., LaForge, R. W., Locander, W. B., MacKenzie, S. B., & Podsakoff, P. M.(2005). New directions in sales leadership research. Journal of Personal Selling & SalesManagement, 25(2), 137–154.

Ingram, T. N., LaForge, R. W., & Schwepker, C. H. (2011). Addressing job stress in thesales force. In D. W. Cravens, K. Le Meunier-FitzHugh, & N. F. Piercy (Eds.). Oxfordhandbook of strategic sales and sales management.

James, L. R., James, L. A., & Ashe, D. K. (1990). The meaning of organizations: The role ofcognition and values. Organizational climate and culture (pp. 40–84). San Francisco:Jossey-Bass.

Jaworski, B. J., & Kohli, A. K. (1993). Market orientation: Antecedents and consequences.Journal of Marketing, 57(3), 53–70.

Johnson, J. S., & Sohi, R. S. (2014). The curvilinear and conditional effects of product linebreadth on salesperson performance, role stress, and job satisfaction. Journal of theAcademy of Marketing Science, 42(1), 71–89.

Jones, E., Brown, S. P., Zoltners, A. A., & Weitz, B. A. (2005). The changing environmentof selling and sales management. The Journal of Personal Selling & Sales Management,25(2), 105–111.

Joshi, A. W., & Campbell, A. J. (2003). Effect of environmental dynamism on relationalgovernance in manufacturer-supplier relationships: A contingency framework and anempirical test. Journal of the Academy of Marketing Science, 31(2), 176–188.

Kim, Y., & Ployhart, R. E. (2014). The effects of staffing and training on firm productivityand profit growth before, during, and after the great recession. Journal of AppliedPsychology, 99(3), 361–389.

Koys, D. J. (2001). The effects of employee satisfaction, organizational citizenship be-havior, and turnover on organizational effectiveness: A unit-level, longitudinal study.Personnel Psychology, 54, 101–114.

Kozlowski, S. W. J., & Klein, K. J. (2000). A multilevel approach to theory and research inorganizations: Contextual, temporal, and emergent processes. In K. J. Klein, & S. W. J.Kozlowski (Eds.). Multilevel theory, research and methods in organizations: foundations,extensions, and new directions (pp. 3–90). San Francisco, CA: Jossey-Bass.

Kumar, V., & Pansari, A. (2016). Competitive advantage through engagement. Journal ofMarketing Research, 53(4), 497–514.

Kumar, V., Sunder, S., & Leone, R. P. (2014). Measuring and managing a salesperson'sfuture value to the firm. Journal of Marketing Research, 51(5), 591–608.

Le Meunier-FitzHugh, K., Massey, G. R., & Piercy, N. F. (2011). The impact of alignedrewards and senior manager attitudes on conflict and collaboration between sales andmarketing. Industrial Marketing Management, 40(7), 1161–1171.

Ledingham, D., Kovac, M., & Simon, H. L. (2006). The new science of sales force pro-ductivity. Harvard Business Review, 84(9), 124–133.

Lenk, P. J., DeSarbo, W. S., Green, P. E., & Young, M. R. (1996). Hierarchical Bayesconjoint analysis: Recovery of partworth heterogeneity from reduced experimentaldesigns. Marketing Science, 15(2), 173–191.

Mallin, M. L., & Pullins, E. P. (2009). The moderating effect of control systems on therelationship between commission and salesperson intrinsic motivation in a customeroriented environment. Industrial Marketing Management, 38(7), 769–777.

Mantel, S. P. (2005). Choice or perception: How affect influences ethical choices amongsalespeople. Journal of Personal Selling & Sales Management, 25(1), 43–55.

Martin, S. W. (2015). What separates the strongest salespeople from the weakest. HarvardBusiness Review. (Available at) https://hbr.org/2015/03/what-separates-the-strongest-salespeople-from-the-weakest (accessed June 6 2018) .

Miao, C. F., & Evans, K. R. (2012). Effects of formal sales control systems: A combinatoryperspective. International Journal of Research in Marketing, 29(2), 181–191.

Mowday, R. T., Porter, L. W., & Steers, R. M. (1979). The measurement of organizationalcommitment. Journal of Vocational Behavior, 14(2), 224–247.

Mulki, J. P., & Jaramillo, F. (2011). Workplace isolation: Salespeople and supervisors inUSA. International Journal of Human Resource Management, 22(04), 902–923.

Mulki, J. P., Locander, W. B., Marshall, G. W., Harris, E. G., & Hensel, J. (2008).Workplace isolation, salesperson commitment, and job performance. Journal ofPersonal Selling and Sales Management, 28(1), 67–78.

Nakata, C., Zhu, Z., & Izberk-Bilgin, E. (2011). Integrating marketing and informationservices functions: A complementarity and competence perspective. Journal of theAcademy of Marketing Science, 39(5), 700–716.

Ostroff, C. (1992). The relationship between satisfaction, attitudes, and performance: Anorganizational level analysis. Journal of Applied Psychology, 77(6), 963–974.

Ovide, S. (2012). Groupon staff feel the heat; some sales reps look to move on as work-load, pay shift. available at http://www.wsj.com/articles/SB10000872396390444900304577581661236440948, Accessed date: 14 September2016.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

16

Peterson, C., Park, N., & Sweeney, P. J. (2008). Group well-being: Morale from a positivepsychology perspective. Applied Psychology, 57, 19–36.

Plouffe, C. R., Bolander, W., Cote, J. A., & Hochstein, B. (2016). Does the customer mattermost? Exploring strategic frontline Employees' influence of customers, the internalbusiness team, and external business partners. Journal of Marketing, 80(1), 106–123.

Power, B. (2016). Why John Deere measures employee morale every two weeks. availableat https://hbr.org/2016/05/why-john-deere-measures-employee-morale-every-two-weeks, Accessed date: 18 November 2016.

Rosen, C. C., Levy, P. E., & Hall, R. J. (2006). Placing perceptions of politics in the contextof the feedback environment, employee attitudes and job performance. Journal ofApplied Psychology, 91(1), 211–220.

Rouziès, D., & Hulland, J. (2014). Does marketing and sales integration always pay off?Evidence from a social capital perspective. Journal of the Academy of MarketingScience, 42(5), 511–527.

Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to jobattitudes and task design. Administrative Science Quarterly, 224–253.

Schein, E. (1992). Organisational culture and leadership. San Francisco: Jossey Bass.Schmitz, C., & Ganesan, S. (2014). Managing customer and organizational complexity in

sales and organizations. Journal of Marketing, 78(6), 59–77.Schneider, B. (1975). Organizational climates: An essay. Personnel Psychology, 28(4),

447–479.Schneider, B., & Rentsch, J. (1987). The management of climate and culture, a future

perspective. Futures of organizationsLexington, MA: Lexington Books.Schulte, M., Ostroff, C., Shmulyian, S., & Kinicki, A. (2009). Organizational climate

configurations: Relationships to collective attitudes, customer satisfaction, and fi-nancial performance. Journal of Applied Psychology, 94(3), 618.

Singh, J., Goolsby, J. R., & Rhoads, G. K. (1994). Behavioral and psychological con-sequences of boundary spanning burnout for customer service representatives.Journal of Marketing Research, 31(4), 558–569.

Sorescu, A. B., Chandy, R. K., & Prabhu, J. C. (2003). Sources and financial consequences

of radical innovation: Insights from pharmaceuticals. Journal of Marketing, 67(4),82–102.

Subramony, M., Krause, N., Norton, J., & Burns, G. N. (2008). The relationship betweenhuman resource investments and organizational performance: A firm-level ex-amination of equilibrium theory. Journal of Applied Psychology, 93(4), 778–788.

Sujan, H., Weitz, B. A., & Kumar, N. (1994). Learning orientation, working smart, andeffective selling. Journal of Marketing, 58(3), 39–52.

Sunder, S., Kumar, V., Goreczny, A., & Maurer, T. (2017). Why do salespeople quit? Anempirical examination of own and peer effects on salesperson turnover behavior.Journal of Marketing Research, 54(3), 381–397.

Tice, T. E. (1997). Managing compensation caps in key accounts. Journal of PersonalSelling & Sales Management, 17(4), 41–47.

Verbeke, W., Dietz, B., & Verwaal, E. (2011). Drivers of sales performance: A con-temporary meta-analysis. Have salespeople become knowledge brokers? Journal ofthe Academy of Marketing Science, 39(3), 407–428.

Whitman, D. S., Van Rooy, D. L., & Viswesvaran, C. (2010). Satisfaction, citizenship be-haviors, and performance in work units: A meta-analysis of collective construct re-lations. Personnel Psychology, 63(1), 41–81.

Wooldridge, J. M. (2009). On estimating firm-level production functions using proxyvariables to control for unobservables. Economics Letters, 104(3), 112–114.

Wright, P. M., Gardner, T. M., Moynihan, L. M., Park, H. J., Gerhart, B., & Delery, J. E.(2001). Measurement error in research on human resources and firm performance:Additional data and suggestions for future research. Personnel Psychology, 54(4),875–901.

Zoltners, A. A., Sinha, P., & Lorimer, S. E. (2006). Match your sales force structure to yourbusiness life cycle. Harvard Business Review, 84(7/8), 80.

Zoltners, A. A., Sinha, P., & Lorimer, S. E. (2008). Sales force effectiveness: A frameworkfor researchers and practitioners. The Journal of Personal Selling & Sales Management,28(2), 115–131.

N.G. Panagopoulos et al. Industrial Marketing Management xxx (xxxx) xxx–xxx

17